Method and control device for controlling motor vehicles

Abstract

This invention proposes a method for automatically controlling a motor vehicle (10) traveling on a road (12) in a current lane (14), wherein the road (12) has at least one additional lane (16). The method includes the steps of: generating and / or receiving at least two temporary driving operations, which include: a transition from the current lane (14) to at least one additional lane (16) and a start time point of the transition. The start times of the at least two temporary driving operations are at different time points. The at least two driving operations are compared, taking into account the corresponding start times points. One of these start times points is selected based on the comparison. Furthermore, a control device for a system used to control a motor vehicle is proposed.

Description

Technical Field

[0001] The present invention relates to a method for controlling a motor vehicle, a control device for a system for controlling a motor vehicle, a motor vehicle, and a computer program for performing the method. Background Technology

[0002] For driver assistance systems that partially automate the longitudinal and lateral movements of motor vehicles, and especially for fully automated motor vehicles, one of the main challenges is analyzing the specific conditions of the vehicle and deriving corresponding driving operations that are meaningful to the vehicle.

[0003] The computational complexity of driving operations typically increases with the duration of each driving operation. If different possible driving operations need to be determined for a longer time period (e.g., longer than three seconds) or if complex driving operations involving multiple lane changes are involved, then previously known methods are often no longer able to acquire these driving operations in real time.

[0004] In almost every traffic condition, there are numerous different possibilities for how to control a motor vehicle. These different possibilities can vary significantly from one another, for example, in terms of required travel time. Especially in heavy traffic, previously known methods are often no longer sufficient to select the appropriate possibility from the various possibilities for controlling the motor vehicle.

[0005] A particular challenge here is the lane change that motor vehicles must make to overtake other road users. The timing of the lane change in particular is crucial, as all subsequent driving maneuver planning depends on it. Current methods are no longer able to determine the optimal starting point for the lane change. Summary of the Invention

[0006] The object of the present invention is therefore to provide a method and a control device for a system used to control motor vehicles, the method and the control device allowing for reliable acquisition of the optimal lane change timing.

[0007] The objective of the invention is achieved by a method for automatically controlling a motor vehicle traveling on a road in a current lane, wherein the road has at least one additional lane. The method includes the steps of: generating and / or receiving at least two temporary driving operations, these temporary driving operations comprising: a change from the current lane to at least one additional lane and a start time point of the change, wherein the start time points of the at least two temporary driving operations are at different time points; comparing the at least two driving operations while considering the corresponding start time points; and selecting one of these start time points based on the comparison.

[0008] The present invention is based on the following basic idea: to obtain the optimal starting time for changing lanes from the current lane to at least one other lane by comparing different temporary driving operations with different predetermined starting times.

[0009] Therefore, the starting time is treated as a parameter that is fixed for each individual temporary driving operation. By comparing these temporary driving operations with different starting time points, the optimal starting time can be determined; or more precisely, the optimal starting time can be selected from multiple different starting time points.

[0010] Here and in the following text, "optimal starting point" should always be understood as those starting points that best meet the predetermined comparison criteria among a number of different starting points.

[0011] In this context and below, "temporary driving operation" should be understood as the motor vehicle's corresponding spatial-temporal trajectory not yet being fixed on a line, but only on a defined spatial-temporal region.

[0012] One aspect of the invention proposes that, in order to compare the at least two driving operations, the following steps are performed:

[0013] - Generate a target driving operation based on one of the at least two temporary driving operations respectively;

[0014] - Compare the at least two target driving operations; and

[0015] - Select one of the at least two target driving operations based on the comparison, wherein the start time point of the temporary driving operation based on the selected target driving operation is selected as the start time point.

[0016] That is, in order to obtain the optimal starting time, instead of simply comparing the corresponding temporary driving operations with each other, we first obtain the target driving operations respectively and then compare these target driving operations with each other.

[0017] The target driving operations involve the following driving actions: when a lane change begins at the corresponding time point, the vehicle should perform the driving operation. Correspondingly, the target driving operations include the spatial-temporal trajectory describing the planned path of the vehicle.

[0018] This method compares the actual final possibilities for trajectory planning of motor vehicles and obtains the optimal starting time point based on this comparison.

[0019] According to another aspect of the invention, at least two temporary driving operations are optimized to generate a target driving operation. In other words, the target driving operation is a set of optimized temporary driving operations.

[0020] Therefore, for multiple possible starting time points, and especially for all possible starting time points, the optimal target driving operation is obtained for each. Each of the obtained target driving operations is "locally optimal," meaning it is optimal given that its corresponding starting time point is fixed. By comparing the optimal target driving operations, the "globally optimal" target driving operation can be obtained. Then, the starting time point belonging to this globally optimal target driving operation is selected.

[0021] According to one aspect of the invention, a cost function is determined, which assigns cost factors to the temporary driving operations at least based on corresponding start times. The cost factors are compared to compare the at least two driving operations, particularly by taking an extreme value of the cost function to optimize the at least two temporary driving operations. The target driving operations are those that cause the cost function to take an extreme value with respect to the corresponding start time. In other words, the cost function has at least one local extreme value, particularly a global extreme value, in the corresponding target driving operation. Correspondingly, the target driving operation is one in which the driving maneuver has been optimized relative to the corresponding temporary driving operation.

[0022] In the following text, “optimized driving operation” should always be understood as a driving operation whose assigned space-time trajectory causes the cost function to be at least locally extreme, and especially globally extreme.

[0023] Another aspect of the invention proposes determining cost functions for the longitudinal motion of the motor vehicle and for the lateral motion of the motor vehicle, respectively. These two cost functions may take extreme values ​​sequentially, in parallel, or independently of each other.

[0024] First, the cost function for the longitudinal motion of the motor vehicle is maximized to obtain the optimized longitudinal trajectory of the vehicle. Then, based on the optimized longitudinal trajectory, the cost function for the lateral motion is maximized to obtain the corresponding target driving operation.

[0025] In particular, the cost function should be minimized to obtain the corresponding target driving operation. Since the smaller the cost factor, the more advantageous the corresponding driving operation, the optimal driving operation, i.e., the target driving operation, can be determined in a simple way.

[0026] It should be noted that in other possible definitions of the cost function, such as the cost function multiplied by (-1), the cost function must be maximized in order to obtain the target driving operation. However, the above-chosen definition of the cost function (i.e., the smaller the cost factor, the more advantageous the corresponding driving operation) corresponds to an intuitive understanding of the cost factor.

[0027] Another aspect of the invention proposes comparing the cost factors assigned to the at least two target driving operations. A cost function or cost factor is used as a comparison criterion to determine which target driving operation should be selected given the respective lane change initiation time. Specifically, the target driving operation with the smallest assigned cost factor and thus the initiation time point with the smallest assigned cost factor are selected.

[0028] According to one embodiment of the invention, the cost function depends, in quadratic form, on the corresponding trajectory assigned to the temporary driving operation, particularly on the trajectory of the vehicle's velocity in the longitudinal direction of the road. That is, the cost function is quadratic, and therefore always has at least one local extremum. In other words, the optimization problem always has a solution.

[0029] In particular, the cost function depends on the first time derivative of the vehicle's (longitudinal) trajectory, i.e., on the vehicle's (longitudinal) velocity. Furthermore, the cost function can depend on the second time derivative of the vehicle's (longitudinal) trajectory, i.e., on the vehicle's (longitudinal) acceleration. Additionally, the cost function can depend on the third time derivative of the vehicle's (longitudinal) trajectory, i.e., on the change in the vehicle's (longitudinal) acceleration, also known as jerk.

[0030] Preferably, optimization of at least two temporary driving operations is performed under at least one additional condition. This at least one additional condition may include safety conditions, comfort conditions, and / or feasibility conditions. An example of a feasibility condition is whether the motor vehicle can reach a certain spatial-temporal range based on its maximum acceleration or deceleration. An example of a comfort condition is whether the acceleration in the longitudinal and / or lateral directions exceeds predetermined limits that are empirically considered uncomfortable by vehicle passengers. An example of a safety condition is the minimum distance or speed limit to be maintained from other road users.

[0031] This at least one safety condition includes, in particular, a predetermined spatial safety distance and / or a predetermined temporal safety distance. Here, "temporal safety distance" should be understood as a time period during which, even if the vehicle does not change its state of motion (e.g., does not brake), the vehicle can certainly continue to move without collision from the current point in time. This time period can also be referred to as the "pre-collision time."

[0032] Here, the spatial safe distance always corresponds to the temporal safe distance, which in turn depends on the vehicle's current speed. More precisely, the temporal safe distance is derived from the ratio of the spatial safe distance to the vehicle's current speed.

[0033] Another aspect of the invention proposes that the at least one additional condition is time-varying and / or depends on the corresponding start time. That is, the time-varying additional condition is considered when limiting the cost function, i.e., when optimizing driving operations. Thus, the current road traffic conditions of the vehicle are considered when optimizing the corresponding driving operations. In particular, this takes into account that the current road traffic conditions of the vehicle change with the start time of the lane change. Therefore, in any road traffic condition, matching driving operations for the vehicle can be generated in real time for different start times.

[0034] In other words, the time-varying additional conditions are not static in time, but rather are additional conditions that change over time, especially with the corresponding starting point and describe the current and / or future traffic conditions in which the motor vehicle is located.

[0035] According to one embodiment of the invention, at least the current lane and / or at least one additional lane are transformed into a Frenet-Serret coordinate system. In this coordinate system, each road is straight, so that traffic conditions on each road can be handled in the same way regardless of the actual direction of the road.

[0036] According to another embodiment of the invention, the motor vehicle is controlled, particularly fully automatically, based on a selected start time point. In other words, the motor vehicle is controlled, particularly fully automatically, according to a selected target driving operation.

[0037] According to the invention, this object is also achieved by a control device for a system for controlling a motor vehicle, or a control device for a motor vehicle, wherein the control device is designed to perform the methods described above. The advantages and characteristics of the control device should be referred to the above explanation of the method for automatically controlling a motor vehicle, which also applies to the control device and vice versa.

[0038] According to the invention, this objective is also achieved by a motor vehicle having the aforementioned control device, particularly wherein the motor vehicle has at least one in-vehicle sensor, particularly an acceleration sensor, a steering angle sensor, radar, lidar, and / or a camera. Regarding the advantages and characteristics of the motor vehicle, reference should be made to the above explanation of the method for automatically controlling a motor vehicle, which is equally applicable to this motor vehicle and vice versa.

[0039] According to the present invention, this objective is also achieved by a computer program having a program encoding medium, for executing the steps of the above-described method when the computer program is implemented on a computer or a corresponding computing unit, particularly the computing unit of the aforementioned control device. Regarding the advantages and characteristics of the computer program, reference should be made to the above explanation of the method for automatically controlling a motor vehicle, which also applies to the computer program and vice versa.

[0040] Herein and in the following text, "program code medium" should be understood as computer-executable instructions in the form of program code and / or compiled and / or uncompiled program code modules, which can exist in any programming language and / or machine language. Attached Figure Description

[0041] Further advantages and features of the present invention will become apparent from the following description and accompanying drawings, with reference to the following description and accompanying drawings. In the accompanying drawings:

[0042] - Figure 1 It schematically illustrates the road traffic conditions;

[0043] - Figure 2 A schematic block diagram of a system for controlling a motor vehicle according to the present invention is shown;

[0044] - Figure 3 A flowchart illustrating the steps of the method according to the present invention is shown;

[0045] - Figure 4 (a) and Figure 4 (b) Schematically showing the path before and after the transformation to the FRINA coordinate system; and

[0046] - Figures 5 to 10 Show respectively Figure 3 The illustrations show the various steps of the method according to the present invention. Detailed Implementation

[0047] Figure 1The diagram schematically illustrates road traffic conditions, with motor vehicle 10 traveling on road 12 in the current lane 14. An additional lane 16 extends alongside the current lane 14.

[0048] In addition, on road 12, another first traffic participant 18 and another second traffic participant 20 are also traveling in the current lane 14 or in another lane 16. In the example shown, the other traffic participants 18 and 20 are passenger vehicles, but could also be trucks, motorcycles, or any other traffic participants.

[0049] There is a lane change zone 21 between the current lane 14 and another lane 16, which partially overlaps with the current lane 14 and the other lane 16.

[0050] As shown by dashed lines 22 and 24, another first traffic participant 18 plans to change from the current lane 14 to another lane 16 via lane change zone 21 in the near future, or another second traffic participant 20 plans to change from another lane 16 to the current lane 14 of motor vehicle 10 via lane change zone in the near future. The other traffic participants 18 and 20 indicate this, for example, by using corresponding turn signals.

[0051] also, Figure 1 The diagram shows a coordinate system with a vertical axis and a normal axis, where the vertical axis defines the longitudinal direction L and the normal axis defines the lateral direction N. The origin of the coordinate system is located at the current position of the front end of the motor vehicle 10 in the longitudinal direction L and, when viewed in the longitudinal direction L, at the right edge of the road.

[0052] The special coordinate system, which will also be used below, is a road-fixed coordinate system, meaning that it does not move with the motor vehicle 10. However, of course, any other coordinate system can be used.

[0053] As in Figure 2 As shown, the motor vehicle 10 has a system 26 for controlling the motor vehicle 10. The system 26 includes a plurality of sensors 28 and at least one control device 30.

[0054] Sensor 28 is disposed at the front, rear and / or sides of motor vehicle 10 and is designed to detect the surrounding environment of motor vehicle 10, generate corresponding environmental data and forward this environmental data to control device 30. More specifically, sensor 28 detects information at least regarding the current driving lane 14, another driving lane 16 and other traffic participants 18, 20.

[0055] The sensors 28 are cameras, radar sensors, distance sensors, lidar sensors, and / or any other type of sensor suitable for detecting the surrounding environment of the motor vehicle 10.

[0056] Alternatively or additionally, at least one of the sensors 28 may be designed as an interface for a guidance system at least assigned to the shown section of road 12 and designed to transmit environmental data about road 12 and / or other traffic participants to motor vehicle 10 and / or other traffic participants 18, 20. In this case, sensor 28 may be implemented as a mobile wireless communication module, for example for communicating according to 5G standards.

[0057] Generally, the control device 30 processes environmental data obtained by the sensor 28 and controls the motor vehicle 10 at least partially automatically, and especially fully automatically, based on the processed environmental data. That is, a driving assistance system is implemented on the control device 30, which can control the lateral and / or longitudinal movements of the motor vehicle 10 at least partially automatically, and especially fully automatically.

[0058] For this purpose, the control device 30 is designed to be used, as described below, by means of Figures 3 to 10 Perform the steps of the method described below. More specifically, the control device 30 includes a data carrier 32 and a computing unit 34, wherein a computer program is stored on the data carrier 32, the computer program is implemented on the computing unit 34 and includes a program encoding medium to perform the steps of the method described below.

[0059] First, the image of road 12, more precisely the current lane 14 and the other lane 16, based on environmental data obtained by sensor 28, is transformed into the Flyner coordinate system (step S1).

[0060] Step S1 in Figure 4 As shown in the image. Figure 4 (a) Road 12 is shown with its actual orientation. In the example shown, viewed in the longitudinal direction L, the road curves to the left. Through a local coordinate transformation, road 12 is transformed to the Flyner coordinate system, where road 12 no longer has a curve, and the result of this transformation is... Figure 4 As shown in (b), road 12 runs straight in this coordinate system and has no curvature along the longitudinal direction L.

[0061] Next, acquire the free area B in the current driving lane 14 and the other driving lane 16. f and occupied area B b (Step S2), where the free area B f and occupied area Bb These are time-space regions.

[0062] Here, in the free area B f The following spatial-temporal zones are defined as follows: These spatial-temporal zones are free from other traffic participants 18, 20 and other obstacles that prevent passage through the corresponding lanes 14, 16.

[0063] Conversely, occupying area B b The following spatial-temporal areas are occupied by other traffic participants 18, 20 and / or by other obstacles, thus making the occupied area B b It cannot be passed by motor vehicle 10.

[0064] To obtain the occupied area, control device 30 requires the predicted trajectories 22, 24 of other traffic participants 18, 20. Control device 30 can determine the trajectories 22, 24 themselves, for example, based on environmental data obtained by sensor 28, such as information that the turn signals of other traffic participants 18, 20 are enabled; or based on data exchanged through inter-vehicle communication. Alternatively, control device 30 can obtain the trajectories 22, 24 directly from other traffic participants 18, 20 or from the guidance system.

[0065] As in Figure 5 China with the help of Figure 1 As shown in the specific example, firstly, the free area B is obtained for the current driving lane 14 and for the other driving lane 16 respectively. f and occupied area B b In particular, the tL curves are used, where t is time.

[0066] In this example, another first traffic participant 18 initiates a lane-changing operation from the current lane 14 to the other lane 16 at time t = 1s, and the lane-changing operation is completed at time t = 5s. Figure 5 In the curve diagram shown, the other first traffic participant 18 occupies both occupied areas B respectively. b The upper part of the occupied area. During the lane change process, another first traffic participant 18 at least temporarily occupies both lanes 14 and 16.

[0067] Another second traffic participant 20 begins a lane-changing operation from another lane 16 towards the current lane 14 at time t = 3s, and completes the lane-changing operation at time t = 7s. Figure 5 In the curve diagram shown, another second traffic participant 20 occupies both occupied areas B respectively. b The lower part of the occupied area.

[0068] Occupied Area Bb The slope here corresponds to the speed of the other traffic participant, either 18 or 20. That is, in Figures 5 to 10 In the example shown, the speeds of the other traffic participants 18 and 20 are constant.

[0069] For simplification, the coordinates in the lateral direction N are discretized, meaning they can only take three different values ​​corresponding to the current lane 14, another lane 16, or lane change zone 21. That is, in Figure 5 The three curves shown are the tL curves for the current driving lane 14, the other driving lane 16, and the lane change zone 21, respectively.

[0070] Here, the shaded sections in the graph correspond to the occupied areas B of lanes 14 and 16, respectively. b Conversely, the unshaded sections in the curve correspond to the open areas B of lanes 14 and 16. f .

[0071] To determine the free area B f First, for each lane 14 and 16, a space-time polygon P is determined. 14 Or P 16 This space-time polygon corresponds to the entire driving lane 14 or 16 in front of the motor vehicle 10, especially the portion of driving lanes 14 or 16 that falls within the range of sensor 28. Figure 5 In the middle, polygon P 14 and P 16 It is a quadrilateral shown by dashed lines.

[0072] In addition, spatial-temporal polygons P are obtained for these two lanes 14 and 16 respectively. 14，b or P 16，b These polygons enclose the occupied areas B of the corresponding driving lanes 14 and 16. b .

[0073] Then, the free area B in the current driving lane 14 is obtained through polygon clipping. f Or more precisely, it corresponds to the free area B. f polygon P 14，f The method is to divide polygon P 14，b From polygon P 14 Remove from the middle. In other words, this involves an operation.

[0074] P 14，f =P 14 \P 14，b .

[0075] Similarly, the empty area B in the other lane 16 is obtained by polygon clipping. f The method is to divide polygon P 16，b From polygon P 16 Remove from the middle. That is, perform operation P. 16，f =P 16 \P 16，b .

[0076] Next, as in Figure 6 As shown in the diagram, the clear sub-region of lane change zone 21 is determined (step S3). Here, lane change zone 21 is clear only when the current driving lane 14 and the other driving lane 16 are clear and lane change zone 21 is not passable for any other reason (e.g., due to an obstacle or due to a no-overtaking rule).

[0077] Therefore, as two polygons P 14，f and P 16，f The intersection of the polygons is used to obtain the free sub-region of lane change area 21, or more precisely, the polygon P corresponding to the free sub-region of lane change area 21. 21，f If lane change zone 21 is impassable due to obstacles or otherwise, then the corresponding spatial-temporal polygon P of the impassable sub-region surrounding lane change zone 21 is... h It is acquired and removed from the intersection mentioned above.

[0078] In other words, the free sub-region P of lane change zone 21 is obtained through operation. 21，f .

[0079] P 21，f =(P 14，f ∩P 16，f )\P h .

[0080] The curves used for the current lane 14 and for the other lane 16 are now divided into time bars (step S4), with a new time bar starting at each event. Figure 7 In the graph, different time bars are separated from each other by vertical dividing lines E, which are inserted into the graph for each event. Here and below, an event should be understood as any type of change in the occupancy status of the corresponding lanes 14 and 16.

[0081] That is, if the occupation of any sub-area of ​​the current lane 14 or another lane 16 begins or ends at a certain point in time, then a new time bar will start at that point in the curve graph used for the current lane 14 or the other lane 16.

[0082] In addition, the dividing line E between the time bars in the curves used for the two lanes 14 and 16 will be transferred to the curve used for the lane change zone 21.

[0083] To achieve consistent layout across the three lane diagrams used for the current lane 14, the additional lane 16, and the lane change zone 21, sloping dividing lines T are inserted into the current lane 14 and the additional lane 16 diagrams, forming the respective occupied areas B. b The extended line of an occupied area. These additional sloping dividing lines T in Figures 8 to 10 As shown in the image.

[0084] Vertical dividing line E, slanted dividing line T, and occupied area B b Divide each of these three curves into multiple sub-regions T. i Where i is a natural number greater than zero, and can take values ​​from 1 to the subregion T. i The total value.

[0085] At least one of these slanted dividing lines and / or at least one of these vertical dividing lines can be omitted or removed again to simplify the following steps of the method. In particular, all slanted dividing lines can be omitted.

[0086] As in Figure 8 The diagram shown below shows the sub-region T of the curve for the current driving lane 14 and for the additional driving lane 16. i Each sub-region in the diagram is assigned a lane vertex V. i And for each sub-region T of the curve graph of lane change zone 21 i Each of these is assigned a lane change zone vertex W. i (Step S5). Here, i is further a natural number greater than zero, and can take values ​​from 1 to the subregion T. i The total value.

[0087] exist Figure 8 In the middle, the apex of the lane V i and lane change zone apex W i Sort the graphs by time, i.e., by comparing them with the sub-regions T with shorter time intervals. i The corresponding vertices are assigned to subregions T with a longer time frame than the vertices assigned to them. i Those vertices are further to the left.

[0088] Next, when the lane apex V is assigned i subregion T i When the driving operation of motor vehicle 10 can be performed between them, the lane apex V of the current driving lane 14 iBy connecting edges in pairs (step S6), more precisely by connecting directed edges in pairs.

[0089] Here, exactly when these two sub-regions T i When they are directly adjacent to each other, that is, when there is no occupied area B. b Driving operations are defined as "possible" only when they are separated from each other. In addition, driving operations can only be performed in the positive time direction.

[0090] For the other lane 16, lane apex V i And lane change zone vertices W of lane change zone 21 i Repeat the same process.

[0091] It should be pointed out that, for clarity, in Figure 9 and Figure 10 The letters "T", "V", and "W" have been omitted. Instead, sub-regions and vertices are simply assigned corresponding numbers. That is, in Figure 9 and Figure 10 In this diagram, the numbers are not part of the attached diagram; rather, they represent the labels of the corresponding sub-regions or vertices.

[0092] Figure 9 The results of step S6 are shown. The graph obtained in step S6 includes all possible driving operations of the motor vehicle 10 within the two driving lanes 14 and 16 and within the lane change zone 21.

[0093] Next, the following lane vertices V of the current driving lane 14. i With the following lane change zone vertex W i Via directed edge connections: the current driving lane 14 or lane change zone 21 is assigned to the subregions T of these lane vertices and these lane change zone vertices. i They overlap (step S7). In other words, if the curves of the current driving lane 14 and lane change zone 21 are superimposed, then the following lane vertex V... i With the following lane change zone vertex W i Connection: Subregion T assigned to these lane vertices and these lane change zone vertices i They have a non-empty intersection.

[0094] In addition, the following lane change zone vertex W i With the following lane apex V of the other driving lane 16 i Via directed edge connection: lane change zones 21 or other lanes 16 are assigned to the vertices of these lane change zones and the sub-regions T of these lane vertices. i They overlap. That is, if the curves of lane 16 and lane change zone 21 are superimposed, then the following lane vertex V... i With the following lane change zone vertex Wi Connection: Subregion T assigned to these lane vertices and these lane change zone vertices i They have a non-empty intersection.

[0095] In other words, in step S7, the free area B is... f Each subregion T i It is divided into a lane-changing area (where lane changes are allowed between lanes 14 and 16) and a lane-keeping area (where lane changes are not allowed between lanes 14 and 16).

[0096] Figure 10 The result of step S7 is shown below. The graph obtained in step S7 contains all possible driving operations of the motor vehicle 10, including changing lanes from the current lane 14 to another lane 16. Here, each of the possible driving operations corresponds to Figure 10 The diagram shows a continuous string of edges.

[0097] Steps S1 to S7 above represent the following possibilities: obtaining many different possible driving operations for the motor vehicle 10, especially all possible driving operations.

[0098] However, it should be noted that any other suitable method can be used to obtain possible driving operations.

[0099] The different possible driving operations thus obtained are then further processed by the optimization module 36 of the control device 30 or by the optimization module of the computer program.

[0100] Generally, the optimization module is designed to obtain the optimal starting time T for lane changing for motor vehicle 10. LC，opt .

[0101] First, a temporary driving operation is generated for each of the multiple possible start time points (step S8).

[0102] For generating multiple temporary driving operations, it is sufficient to generate a separate space-time trajectory for each of these possible starting points, for example, starting from the lane-assigned vertex V. i Or the apex of the lane change zone W i Individual points are selected in each spatial-temporal region and then these points are connected to each other. However, the resulting trajectory should be smooth, i.e., without inflection points. For example, the points can be connected to each other using spline functions to obtain the spatial-temporal trajectory.

[0103] Given the various possibilities of temporary driving operations at a given starting time, we can propose selecting the temporary driving operation with the shortest assigned trajectory.

[0104] That is, the generated temporary driving operations are each connected to a specific starting point of time for lane changing for motor vehicles. In the temporary driving operation, the corresponding spatial-temporal trajectory of the motor vehicle is not yet fixed on a line, but is only limited to a defined spatial-temporal region.

[0105] Now optimize the temporary driving operations separately, so that a target driving operation is obtained from each of these temporary driving operations (step S9).

[0106] To optimize temporary driving operations, a cost function F is determined, which is a space-time trajectory. The assignment cost factor K = F(x) is a space-time trajectory describing the starting time T for a lane change by a motor vehicle. LC，i The corresponding driving operations.

[0107] Cost functions can include, for example, the following forms:

[0108]

[0109] Here, L E It is the (longitudinal) trajectory of motor vehicle 10. It is the expected final speed of the motor vehicle 10 and T LC，pre This corresponds to the temporary lane change start time. Coefficient γ i The weighting factor represents the weighting factor of each term in the cost function.

[0110] That is, in this example, the cost function explicitly depends on the speed of vehicle 10, the acceleration of vehicle 10, and the expected final speed of vehicle 10. Conversely, the dependence on the starting time point is only implicit.

[0111] Another possibility for the cost function is as follows:

[0112]

[0113] Here, the cost function additionally depends on the third-order time derivative of the vehicle's (longitudinal) trajectory, i.e., on the jerk.

[0114] Furthermore, due to the term γ3T LC，pre The cost function explicitly (i.e. directly) depends on the starting time of the lane change. Therefore, it can be assumed that, where appropriate, the vehicle 10 will want to change lanes as quickly as possible, for example, to avoid its own braking.

[0115] To optimize the corresponding temporary driving operations, the cost function is optimized by finding its extreme value, or more precisely, minimizing it. That is, for multiple starting time points, the following driving operations are determined: their spatial-temporal trajectories cause a local minimum in the cost function. The optimized driving operation, i.e., the corresponding target driving operation, is thus assigned a locally minimum cost factor, and especially a globally minimum cost factor.

[0116] Motor vehicle 10 is subject to various inherent limitations. For example, motor vehicle 10 has maximum acceleration and maximum deceleration that cannot be exceeded, respectively. In addition, motor vehicle 10 is subject to various external limitations. For example, motor vehicle is not allowed to remain in the same place as an obstacle, as this would be equivalent to a collision. Furthermore, most roads have speed limits.

[0117] To account for both inherent and external constraints, additional conditions are derived based on these constraints, and the cost function is then minimized under these additional conditions. In this way, a target driving operation is obtained, which is the optimal driving operation among the corresponding driving operation categories under the given additional conditions.

[0118] At least one of the additional conditions may be a time-varying additional condition that is determined individually for the corresponding driving operation category (step S10).

[0119] The at least one additional condition that varies with time is obtained based on the space-time polygons acquired in steps S1 to S5. These space-time polygons describe the free space-time region B. f and the space-time region B occupied b and subregion T i .

[0120] Motor vehicles 10 are not permitted to remain in these occupied areas B at any time. b Within an occupied area. However, since a certain safety distance is usually required, this condition alone is often too weak as an additional requirement.

[0121] Therefore, the at least one additional condition that varies over time is determined in such a way that it includes the occupied area B. b A predetermined safe distance in time and / or a predetermined safe distance in space.

[0122] Here, "safe distance in time" should be understood as a time period T. TTC During this time period, even if the vehicle does not change its state of motion (e.g., does not brake), the vehicle can still move without collision from the current point in time. This time period can also be called the "pre-collision time".

[0123] Here, the spatial safety distance always corresponds to the temporal safety distance, which in turn depends on the vehicle's current speed. More precisely, the temporal safety distance is derived from the quotient of the spatial safety distance and the vehicle's current speed v(t).

[0124] Here, the at least one additional condition that changes over time can typically be formulated in this manner.

[0125]

[0126] That is, for any given time point t, the longitudinal coordinate L of the space-time trajectory x(t) has a distance from the occupied region B. b The minimum distance, where the minimum distance depends on the current velocity v(t).

[0127] Here, the additional conditions that change over time also depend on the exact location of the vehicle 10. For example, if the vehicle is located in sub-region T6 (see...) Figure 8 If the condition is L(t) + v(t)·T, then the corresponding additional condition is L(t) + v(t)·T. TTC ≤max P 14，b (t+T TTC ).

[0128] It should be noted that the above inequality should be understood conceptually rather than literally. “max P” should be understood here as the corresponding polygon constraining the sub-region T5 on the upper side.

[0129] To put it figuratively, even if motor vehicle 10 is in time T TTC The vehicle continues its motion at a constant speed within the polygon P, and is not allowed to be in a position within the polygon P. 14，b Within a defined area.

[0130] Conversely, if the motor vehicle is in subregion T 15 In this context, the additional conditions that change over time are as follows:

[0131] min P 16，b (t+T_TTC)≤L(t)+v(t·T) TTC ≤max P 14，b (t+T TTC )

[0132] Similar to the above situation, min P 16，b (t+T TTC This describes the lower-side restricted subregion T. 15 Polygons.

[0133] That is, for each sub-region T iIndividually determine the time-varying additional conditions for optimizing driving operations. Correspondingly, these time-varying additional conditions can also change across the entire space-time trajectory.

[0134] Especially based on the corresponding start time point T LCi This is to obtain at least one additional condition that varies over time. Correspondingly, the additional condition that varies over time can depend on the starting time point T. LC .

[0135] Additional conditions can also be set, such as safety conditions, comfort conditions, and / or feasibility conditions. An example of a feasibility condition is whether the vehicle can reach a certain spatial-temporal range based on its maximum acceleration or deceleration. An example of a comfort condition is whether the acceleration in the longitudinal and / or lateral directions exceeds predetermined limits that are empirically considered uncomfortable by vehicle passengers. An example of a safety condition is the minimum distance or speed limit to be maintained from other road users.

[0136] The target operation can then be determined, for example, by applying a triple integrator, which is defined as follows:

[0137]

[0138]

[0139]

[0140] L E (0)=L E，0 ,

[0141] Where h is a positive real number greater than zero.

[0142] In summary, in step S9, conceptually different driving operations, i.e., temporary driving operations, are optimized under determined additional conditions. Correspondingly, for each temporary driving operation, the optimal driving operation is obtained, i.e., the corresponding target driving operation is obtained.

[0143] That is, the result of step S9 is a set of optimal driving operations belonging to different lane change start times.

[0144] It should be noted that the optimization can be carried out in two stages, where the longitudinal trajectory of the vehicle 10 is optimized first, and then the lateral trajectory of the vehicle is optimized.

[0145] Now, multiple target driving operations or corresponding cost factors assigned to target driving operations are compared with each other, and based on the comparison, one of these target driving operations is selected as the target driving operation and thus one of these start time points is also selected (step S11).

[0146] More precisely, the target driving operation with the lowest assigned cost factor is selected.

[0147] The motor vehicle 10 can then be controlled by the control device 30, at least partially automatically, and especially fully automatically, according to the selected driving operation.

[0148] Based on the above method, the optimal starting time T for changing lanes from the current lane 14 to at least one other lane 16 is obtained in the following manner. LC，opt : ... LC，i Different temporary driving operations are compared with each other. To this end, optimized target driving operations are obtained based on the temporary driving operations.

[0149] Here, the starting time point T is treated as a parameter that is fixed for each individual temporary driving operation and for each target driving operation. LC，i By comparing the cost factors assigned to the target driving operation, the optimal starting time T can be determined. LC，opt Or more precisely, it can be from multiple different starting time points T LC，i The optimal starting time point is selected from among them.

[0150] List of reference numerals

[0151] 10 Motor vehicles

[0152] 12 roads

[0153] 14 Current driving lane

[0154] 16 Other driving lanes

[0155] 18 Other first traffic participants

[0156] 20 Other second traffic participants

[0157] 21 Lane Change Zone

[0158] 22 Dashed line

[0159] 24 Dashed lines

[0160] 26 Systems for controlling motor vehicles

[0161] 28 sensors

[0162] 30 Controllers

[0163] 32 Data carriers

[0164] 34 Computing Units

[0165] 36 Optimization Mode

Claims

1. A method for automatically controlling a motor vehicle (10) traveling in a current lane (14) on a road (12), wherein the road (12) has at least one additional lane (16), the method comprising the steps of: - Generate and / or receive at least two temporary driving operations, the driving operations including: The change from the current lane (14) to the at least one other lane (16) and the starting time of the change, wherein the starting time of the at least two temporary driving operations is at different time points; - Determine a cost function that assigns cost factors to the temporary driving operations based at least on corresponding start times, wherein the cost factors are compared to compare at least two driving operations, and wherein the cost function is extreme to optimize the at least two temporary driving operations. To compare at least two driving operations, the following steps are performed: - Generate a target driving operation based on one of the at least two temporary driving operations, wherein the at least two temporary driving operations are optimized to generate the target driving operation; - Compare the at least two target driving operations; and - Select one of the at least two target driving operations based on the comparison, wherein the start time point to which the temporary driving operation based on the selected target driving operation belongs is selected as the start time point.

2. The method according to claim 1, characterized in that, The cost factors assigned to the at least two target driving operations are compared with each other in order to compare the at least two target driving operations.

3. The method according to claim 1 or 2, characterized in that, The cost function depends in quadratic form on the corresponding trajectory of the temporary driving operation assigned to it.

4. The method according to claim 3, characterized in that The at least two temporary driving operations are optimized under at least one additional condition.

5. The method according to claim 4, characterized in that, The at least one additional condition is time-varying and / or depends on the corresponding start time.

6. The method according to claim 3, characterized in that, At least the current lane (14) and / or at least one other lane (16) are transformed into the Flener coordinate system.

7. The method according to claim 1, characterized in that, The motor vehicle (10) is controlled based on the selected start time point.

8. The method according to claim 1 or 2, characterized in that, The cost function, in quadratic form, depends at least on the speed of the motor vehicle in the longitudinal direction (L) of the road (12).

9. The method according to claim 1, characterized in that, The motor vehicle (10) is controlled fully automatically based on the selected start time point.

10. A system (26) for controlling a motor vehicle (10) or a control device for a motor vehicle (10), wherein the control device (30) is designed to perform the method according to any one of the preceding claims.

11. A motor vehicle having a control device (30) according to claim 10.

12. The motor vehicle according to claim 11, characterized in that, The motor vehicle (10) has at least one in-vehicle sensor (28).

13. The motor vehicle according to claim 12, characterized in that, The sensors are accelerometers, steering angle sensors, radar, lidar, and / or cameras.

14. A computer program having a program encoding medium, for performing the steps of the method according to any one of claims 1 to 9 when the computer program is implemented on a computer or a corresponding computing unit.

15. The computer program according to claim 14, characterized in that, The computing unit is the computing unit (34) of the control device (30) according to claim 10.

Images