METHOD FOR TRAJECTORY PLANNING OF AN ASSISTANCE SYSTEM

Abstract

The invention relates to a method for trajectory planning of a driver assistance system, particularly an assistance system for longitudinal and/or transverse control, in which a trajectory (T1, T3-T7) having a total duration (t.sub.e) which can be set is determined and the trajectory (T3, T5, T7) is divided into segments, wherein each segment has a variable segment duration (Δt.sub.1, Δt.sub.2, Δt.sub.3) and the total of the segment durations (Δt.sub.1, Δt.sub.2, Δt.sub.3) corresponds to the set total duration (t.sub.e) of the trajectory (T3, T5, T7).

Claims

1. A method for trajectory planning of an assistance system, particularly an assistance system for longitudinal and/or transverse control, in which a trajectory (T1, T3-T7) having a total duration (t.sub.e) which can be set is determined, and the trajectory (T3, T5, T7) is divided into segments, wherein each segment has a variable segment duration (Δt.sub.1, Δt.sub.2, Δt.sub.3) and the total of the segment durations (Δt.sub.1, Δt.sub.2, Δt.sub.3) corresponds to the set total duration (t.sub.e) of the trajectory (T3, T5, T7).

2. The method according to claim 1, characterized in that the trajectory (T3, T5, T7) is divided into segments as a function of the respective acceleration and/or speed.

3. The method according to claim 1 or 2, characterized in that a segment for building the acceleration, a segment for holding and/or altering the acceleration and a segment for reducing the acceleration are provided.

4. The method according to at least one of the preceding claims, characterized in that the accelerations in the first and in the third segments of the trajectory (T3, T5, T7) are in each case described by a higher-order polynomial, particularly a third-order or fifth-order polynomial.

5. The method according to at least one of the preceding claims, characterized in that the segment duration (Δt.sub.1, Δt.sub.2, Δt.sub.3) of the respective segments is set based on a quality measure.

6. The method according to at least one of the preceding claims, characterized in that the segment duration (Δt.sub.1) of the first segment is determined as a function of the segment duration (Δt.sub.3) of the third segment or vice versa.

7. The method according to at least one of the preceding claims, characterized in that a three-part trajectory (T7) is calculated for following a vehicle in that the partial trajectories in the first and second segments of the trajectory (T7) substantially correspond to the partial trajectories of the planning for traveling on a clear section of road (T3, T5), whilst the third segment is described by a polynomial of another order, in particular a fifth-order polynomial, so that the trajectory is converted into a desired final condition for acceleration, speed (v.sub.e) and path.

8. The method according to at least one of the preceding claims, characterized in that the segment duration (Δt.sub.1, Δt.sub.2, Δt.sub.3) of one of the segments is chosen in such a way that the quality measure of the three-part trajectory (T3, T5, T7) becomes minimal.

9. The method according to at least one of the preceding claims, characterized in that a subordinate optimization is provided for selecting the segment duration (Δt.sub.1, Δt.sub.2, Δt.sub.3) of a segment.

10. The method according to at least one of the preceding claims, characterized in that the trajectory (T1, T3-T7) is planned by varying the total duration (t.sub.e) of the respective trajectory.

11. The method according to at least one of the preceding claims, characterized in that an adaptive search space having raster points for predefining trajectory target conditions is provided, and an optimal trajectory (T1, T3-T7) is determined based on a displacement of the raster points, particularly in such a way that the raster points are displaced towards the trajectory to be determined (T1, T3-T7).

12. The method according to claim 11, characterized in that the raster points are adapted iteratively over multiple time steps.

13. The method according to at least one of the preceding claims, characterized in that a spring-damper system is provided for generating target driving conditions of the means of transportation for following a vehicle.

14. The method according to at least one of the preceding claims, characterized in that, for following a vehicle, a spring-damper system is arranged as a virtual bumper between the means of transportation and a means of transportation driving ahead in order to regulate the distance, and the clearance which can be set between the means of transportation, the speed and/or spring travel are enlisted for arranging the virtual bumper.

15. The method according to at least one of the preceding claims, characterized in that at least one acceleration and/or speed plateau is provided for compensating for control deviations.

16. The method according to at least one of the preceding claims, characterized in that a trajectory planner is provided for determining the trajectory (T1-T7).

17. The method according to claim 16, characterized in that the trajectory planner comprises multiple modules and/or levels.

18. The method according to claim 16 or 17, characterized in that the trajectory planner comprises a coordination level (1) for situation-specific and function-specific adjustment of a target condition and a planning level (2) for determining a trajectory based on the target condition.

19. The method according to at least one of the preceding claims, characterized in that the method is implemented as an algorithm.

20. A trajectory planner for an assistance system, comprising a coordination level (1) for predefining a target condition, a planning level (2) for determining a trajectory (T1-T7) based on the target condition, and a trajectory selection module (3) for selecting the respective trajectory (T1-T7), wherein the trajectory planner is designed in such a way that the trajectory planning is performed by means of a method according to at least one of the preceding claims.

21. An assistance system for a means of transportation, particularly for longitudinal and/or transverse control, characterized in that in the case of the assistance system trajectory planning is carried out by means of a method according to any one of claims 1-19.

Description

DESCRIPTION OF THE INVENTION WITH REFERENCE TO EXEMPLARY EMBODIMENTS

[0029] The invention will be explained in greater detail below with reference to expedient exemplary embodiments, wherein:

[0030] FIG. 1 shows a simplified schematic diagram of a configuration of a structure of a trajectory planner according to the invention;

[0031] FIG. 2 shows a simplified diagram of trajectories for traveling on a clear section of road according to the prior art;

[0032] FIG. 3 shows a simplified diagram of a trajectory (dotted) within the meaning of the invention for the one-part trajectory from FIG. 2;

[0033] FIG. 4 shows a further simplified diagram of a trajectory planned according to the invention;

[0034] FIG. 5 shows a further simplified diagram of a trajectory planned according to the invention for distance regulation;

[0035] FIG. 6 shows a simplified diagram of a nominal condition specification when following a vehicle;

[0036] FIG. 7 shows a simplified diagram of a mass-spring-damper system for generating nominal conditions on a vehicle, and

[0037] FIG. 8 shows a simplified diagram of a virtual bumper between an ego vehicle and a vehicle driving ahead.

[0038] Exemplary embodiments according to the invention for calculating multi-part trajectories are described below. A trajectory converts the system condition from its initial value into a defined final value. The system condition is described by the position s, the speed v, the acceleration a and, depending on the system model, by the jolt r. The vehicle is modelled by a point mass for the trajectory calculation. In particular, a multi-step integrator chain is, as a general rule, used as a system model. The trajectory calculation represents an optimization problem which can be solved analytically according to the prior art. However, such solutions, as a general rule, describe the system conditions by polynomials which have the disadvantage that they can only reach the maximum values of jolt and acceleration on a point-by-point basis and cannot be kept constant in sections.

[0039] An exemplary embodiment of a structure of a trajectory planner according to the invention for a driver assistance system is depicted in FIG. 1. Such a trajectory planner can, however, be expressly utilized for assistance systems of other means of transportation (flying objects, watercraft and the like). The trajectory planner comprises a coordination level 1 (or coordination layer) and a planning level 2 (or planning layer). Whilst the planning level 2 universally optimizes and calculates trajectories for converting the vehicle from its current actual condition into a desired target condition, the coordination level 1 provides an interface for situation-specific and function-specific adjustment of the target condition, the optimization criteria and restrictions of the trajectory planning.

[0040] Due to the various optimization goals for traveling on a clear section of road without a target object (speed trajectory) and for following a vehicle with a target object (distance regulation), the planning level 2 consists of a planner for speed trajectories (speed planner 9) and one or more (multi-object ACC) planners for distance trajectories (clearance or route planner 10). So-called traveling on a clear section of road designates the process of a vehicle driving and its own lane being clear or no vehicle driving ahead being established as a relevant target object, meaning that the vehicle is able to drive in an unimpeded manner at a target speed adjusted by the driver. If, however, a vehicle driving ahead which prevents the vehicle traveling on a clear section of road is established by, e.g., an assistance system, the speed can be controlled accordingly and adapted to the speed of the vehicle driving ahead. Correspondingly, this process involves so-called following a vehicle, in which the speed is, as a general rule, adjusted based on a nominal clearance which can be set with respect to the vehicle driving ahead. In order to change between the trajectories for traveling on a clear section of road and following a vehicle, a trajectory selection module 3 is linked to the different planners. The trajectories can be selected on the basis of the current trajectory acceleration. Alternatively, the trajectories can also be selected based on the complete trajectories.

[0041] The coordination level 1 preferably has a modular construction and contains an independent module or multiple independent modules such as, e.g., a speed module 4 and a distance assistance module 5 for each delimitable functionality of the system. Each module provides an intuitive interface with the application of the respective functionality. For this purpose, the respective module translates and reduces the plurality of optimization parameters of the actuated trajectory planner (weightings in the quality measure, condition restrictions, search space boundaries) to a few parameters in order to parameterize the respective functionality in a targeted manner. More complex algorithms are also conceivable, which control the behavior in complete scenarios. The modules can likewise have a modular construction and comprise subordinate functions or modules. As depicted by way of example in FIG. 1, the speed module 4 comprises at least three further (subordinate) modules: a speed control module 6, a speed limit assistance module 7, as well as a cornering assistance module 8. The individual modules consequently offer an intuitive interface with the situation-specific application of the subordinate planners and, thus, of the resulting trajectories or the desired trajectories. The plurality of optimization parameters of the trajectory planning (e.g., weightings in the quality measure, condition restrictions, choice of search space) are not released directly for the application since the function modules of the coordination level 1 initially translate and reduce the application task to a few plausible parameters in order to adjust the desired trajectory behavior in a targeted manner.

[0042] The coordination level 1 further offers the possibility of arbitrating in advance between different functionalities, or the coordination level 1 can assume the arbitration between different functionalities. For example, the requirements and target conditions of functions for controlling the speed without a target object (e.g., on the basis of a driver specification, predictive traffic sign recognition or predictive curve recognition) can be compared in advance so that, e.g., only the most critical requirement for speed trajectories are forwarded to the speed planner 9.

[0043] The arbitration between safety and comfort functions can also be carried out in the same way. For example, an EBA requirement can constantly override an ACC requirement, i.e., the respective functions can be prioritized for safety-critical viewpoints. On the other hand, it may be necessary in the case of distance regulation to calculate multiple planners for distance or clearance trajectories in parallel since multiple target objects are frequently located in the immediate vehicle environment (in front of or on the adjacent lanes) and it is not always the case that the most critical object is known in advance and can be selected for planning. One such example is a scenario in which overtaking on the “slower lane” (“overtaking on the right maneuver”) with target objects in the vehicle's own lane and the adjacent lane is to be prevented. Here, the additional planners for distance trajectories can be designed, if possible, more simply (e.g., by means of a limited/coarser rasterization of the search space) than the main planner which is optimized for maximum comfort, e.g., with an increasing number of relevant objects, in order to limit the resource requirements.

[0044] FIG. 2 shows the speed v (top) and the acceleration a (bottom) of an exemplary one-part trajectory T1 for traveling on a clear section of road according to the prior art. In the present example, the speed is now to be increased from 10 m/s to 20 m/s, wherein an acceleration restriction of 2 m/s.sup.2 is effective. Since the calculated one-part trajectory violates the acceleration restriction, it is classified according to the prior art as inadmissible and discarded. In order to make better use of the accelerating power of the vehicle, three-part trajectories can be deployed. Such a three-part trajectory T2 is depicted in FIG. 2 in addition to the one-part trajectory T1. Therein, the first trajectory segment guides the acceleration to the maximum or minimum value a.sub.cst, the second segment keeps the acceleration constant and the third segment reduces the acceleration again. The duration for the first and third trajectory segments is kept constant and the duration of the second segment is varied in such a way that the desired final speed is achieved. The consequence of this is that the duration t.sub.e of the three-part trajectory generally deviates from the duration of the one-part trajectory T1, as shown in FIG. 2. Comparing the one-part trajectory T1 and the three-part trajectories T2 is thus inconsistent, since the trajectory length is included in the quality measure. A further disadvantage follows from the invariant duration of the first and third trajectory segments which, as a result, cannot be adapted to the specific situation.

[0045] In contrast, the calculation of three-part speed trajectories is proposed according to the invention with a variable duration of all of the trajectory segments while observing the total duration t.sub.e. The duration of the respective segments follows due to the minimization of a quality measure. The quality measure evaluates the deployment of the control variable at the start of the route model or the integrator chain and corresponds in this respect to the integral portion of the quality measure for evaluating one-part trajectories. Due to the consistent trajectory duration and the consistent quality measure of one-part and three-part trajectories, these can be exchanged directly in the superimposed optimization.

[0046] To ensure that a three-part trajectory attains the specified speed change v.sub.e−v.sub.0, the following equation has to be met:

∫.sub.0.sup.Δt.sup.1a.sub.1dt+∫.sub.0.sup.Δt.sup.2a.sub.cstdt+∫.sub.0.sup.Δt.sup.3a.sub.3dt=v.sub.e−v.sub.0.

[0047] The total trajectory length t.sub.e corresponds to the total of the segment durations Δt.sub.1, Δt.sub.2 and Δt.sub.3. The accelerations a.sub.1 and a.sub.3 in the first and in the third segments are described by third-order polynomials. The insertion of a.sub.1 and a.sub.3 in the above equation leads to the quadratic equation for Δt.sub.1

a.Math.Δt.sub.1.sup.2+b.Math.Δt.sub.1+c(Δt.sub.3,t.sub.e)=0,

if Δt.sub.3 and t.sub.e are assumed as parameters, i.e., Δt.sub.1 can be calculated for meaningfully chosen values of Δt.sub.3 and the resulting three-part trajectory has the required length t.sub.e. It has been shown that, with two valid solutions for Δt.sub.1, the smaller results in a smaller quality measure. The duration Δt.sub.3 is chosen so that the quality measure of the three-part trajectory becomes minimal. A subordinate optimization is deployed for this purpose. In a first step, the possible solution range for Δt.sub.3 is determined. Initially, this cannot be less than zero and not longer than the trajectory length t.sub.e. Only real and positive solutions are possible for solving the quadratic equation for Δt.sub.1, which leads to two inequations. A third inequation arises from the further demand that Δt.sub.2 is to likewise be positive. In order to find the optimal Δt.sub.3, a bisection method is deployed in a second step. This produces an optimal replacement for the one-part trajectory, which is consistent with the latter, after a few calculation steps. In FIG. 3, a three-part trajectory T3 within the meaning of the invention is depicted for the example in FIG. 2. In FIG. 4, a further example is depicted, in which a non-symmetrical trajectory T5 is calculated as a replacement for the one-part trajectory T4.

[0048] The problem described above is also relevant in the case of path trajectories. Due to the higher order of the polynomials in path planning, a trajectory can, however, violate both the lower and the upper acceleration restriction. In such a case, the replacement trajectory has up to five trajectory segments and can no longer be calculated analytically. In practice, however, it is much more important that the lower acceleration restriction is exploited (e.g., access scenarios). In the event that only one acceleration restriction is exploited, a three-part path trajectory can be calculated in a similar way to the speed trajectory.

[0049] The first and the second trajectory segments a.sub.1 and a.sub.k are identical to the case of the speed trajectories, while the third segment converts the final condition to the desired final speed v.sub.e by a fifth-order polynomial. To ensure that a three-part trajectory bridges the specified distance s.sub.e-s.sub.0, the following equation has to be met:

∫∫.sub.0.sup.Δt.sup.1a.sub.1dt+∫∫.sub.0.sup.Δt.sup.2a.sub.cstdt+∫.sub.0.sup.Δt.sup.3v.sub.3dt=s.sub.e−s.sub.0.

[0050] After inserting the acceleration a.sub.1 and the speed v.sub.3, this produces the quadratic equation for Δt.sub.2

a.Math.Δt.sub.2.sup.2+b(Δt.sub.1,t.sub.e).Math.Δt.sub.2+c(Δt.sub.1,t.sub.e)=0,

if Δt.sub.1 and t.sub.e are assumed as parameters. In the case of the path trajectories, it no longer has to be decided in advance whether, in the case of two valid solutions for Δt.sub.2, the smaller one also leads to a smaller quality measure. Therefore, both solutions have to be further investigated. The solution range of Δt.sub.1 can be restricted by way of inequations. In FIG. 5, an exemplary embodiment for replacing a one-part path trajectory T6 with a three-part trajectory T7 is depicted (top path, middle speed and bottom acceleration): the three-part trajectory T7 (dotted) within the meaning of the invention stands, by way of example, for a trajectory having

r.sub.0=0 m/s.sup.3,

a.sub.0=0 m/s.sup.2,

v.sub.0=8 m/s,

r.sub.e=0 m/s.sup.3,

a.sub.e=0 m/s.sup.2,

v.sub.e=2 m/s and

s.sub.e=40 m.

An alternative possibility for planning multi-part path trajectories is produced by integrating three-part speed trajectories. A trajectory which approximates a one-part path trajectory, i.e., approaches a one-part trajectory, can be found by way of varying the trajectory end time and evaluating the resulting final clearance.

[0051] Target conditions for the trajectory planning can be further specified. The nominal clearance when following a vehicle can be determined based on the following equation:

d.sub.w=d.sub.stop+v.sub.t.Math.headway.

[0052] Therein, d.sub.stop stands for the clearance with respect to the target vehicle when stationary, v.sub.t stands for the speed of the target vehicle and “headway” stands for the time gap. Furthermore, the movement of the target vehicle can be predicted in future, assuming a constant acceleration a.sub.t,0:

a.sub.t=a.sub.t,0

v.sub.t=v.sub.t,0+a.sub.t,0t

s.sub.t=d.sub.0+v.sub.t,0t+½a.sub.t,0t.sup.2.

[0053] The measured clearance is denoted by d.sub.0. The nominal position s.sub.w of the ego vehicle results from the predicted position of the target vehicle and the nominal clearance, according to:

s.sub.w=s.sub.t−d.sub.w=s.sub.t−d.sub.stop−v.sub.t.Math.headway.

[0054] The remaining nominal conditions v.sub.w (speed) and a.sub.w (acceleration) can be calculated by differentiating this equation, e.g., by

v.sub.w=v.sub.t−a.sub.t.Math.headway

a.sub.w=a.sub.t.

[0055] The courses of the conditions (path or distance (top), speed (middle) and acceleration (bottom)) of a target vehicle and the resulting nominal conditions are depicted in FIG. 6. The condition of the target vehicle is depicted in continuous black and the target conditions for the trajectory planning according to the clearance equation are depicted in dotted-dashed-black and following the filtering with a mass-spring-damper system (dashed line). In the event of a change in the acceleration of the target vehicle, the predicted nominal speed jumps by −a.sub.t* headway. Thus, at the start of a braking of the target vehicle (e.g., from 12 s to 17 s) a higher target speed than that of the target vehicle is rapidly required. As a consequence, the trajectory planning finds solutions which accelerate the vehicle in order to arrive at this higher speed, i.e., the vehicle is accelerated by the nominal clearance which is becoming smaller. Conversely, the nominal speed jumps to zero if the predicted target vehicle comes to a standstill. This jump in the nominal condition specification unfavorably results in the vehicle following a stopping or stationary target vehicle at too high a speed and with a small clearance. When the target vehicle moves off from a standstill, the nominal clearance increases again so that the nominal values for the planning lie at negative speeds for a while and would therefore push the stationary vehicle backwards.

[0056] The insertion or provision of a virtual mass-spring-damper system arranged on the target vehicle represents one configuration of the method (FIG. 7). The nominal clearance d.sub.w corresponds to the spring length l, wherein c describes the spring constant. The condition x.sub.r of the mass m is used as a new nominal condition for the planning. This is a consistent filtering of the target vehicle condition in all conditions. The filtered nominal condition is depicted in FIG. 6. During braking of the target vehicle (from 12 s to 17 s), the target speed rises steadily to the nominal value without a jump. As a result, the specified clearance is slightly greater than the calculated nominal clearance and makes it possible to brake to a standstill, without the nominal speed and the nominal acceleration having inconsistencies. The invention also expressly comprises further configurations or interconnections of springs and dampers, which are not explicitly mentioned. Stability can be achieved in a convoy thanks to such a filtering.

[0057] A dynamic search space can also be expediently provided. How well the exact solution of the optimization problem is approximated by the calculated trajectory is determined by rasterizing the search space. Precise rasterization contrasts with low computing requirements. However, it is necessary in specific situations to rasterize very precisely in order to find a valid solution in the search space.

[0058] A disadvantage of a fixed and coarse rasterization is the non-continuous influence of the optimization parameters on the trajectory found. Parameter changes do not bring about a change in the trajectory until another raster point has a lower quality measure than the current raster point. This behavior makes application more difficult and prevents an intuitive course of action. Therefore, an adaptive search space is proposed. This is understood to mean that the raster points are displaced in such a way that more raster points are located in the vicinity of the best solution than in more remote areas. It is furthermore important that raster points are located in the entire search space so that it is possible to react very quickly to erratic changes in the target conditions (e.g., change in target object, strong target object braking, etc.). By iteratively adapting the raster points over multiple cycles of the optimization, it is possible to refine the rasterization about a selected point and to approach the optimum in this way. However, it can furthermore occur that no valid solution is found. The raster points can be focused statically or dynamically around the current solution. Genetic optimization represents another possibility. Here, solid, free charged and free uncharged particles are deployed. The solid particles delimit the search space, the free particles iterate towards the optimum and the charged particles cover a range around the solution. Here, each time step represents a generation. In a practical way, the particles do not concentrate on one point even after several hundred iterations. A displacement of the optimum which has taken place over time can be further considered during genetic optimization.

[0059] According to a further practical configuration of the invention, a “stop-and-go-function” can be provided, particularly for an ACC control. In a practical way, for such stop-and-go-functionality, the vehicle can follow a vehicle driving ahead right down to a standstill and move off again when the vehicle driving ahead moves off. The stopping process can be conveniently and reproducibly designed by defined “crawling” (i.e., continuing to move particularly slowly) just before coming to a standstill.

[0060] The previously described filtering of the target vehicle conditions generates nominal conditions which are preferably familiar to the driver. While the prediction of such target conditions has a jump to a negative target speed based on unfiltered target vehicle data when following a target vehicle moving off from a standstill, continuous nominal conditions with consistently positive target speed and target path specifications are produced with the present developed filtering of the target vehicle data. The same is also shown for an erratic change in the target vehicle acceleration (positive or negative) when following a vehicle normally (at speeds v>0 km/h).

[0061] When braking right down to a standstill behind a target vehicle it can happen that, due to too coarse a rasterization, there is no suitable target point in the quantity of target points. Too early a time requires stronger braking and too late a time leads to short-term reversing. Correspondingly, constantly varying target vehicle data and the resulting adaptation of the trajectories can lead to time steps, in which no suitable solution is found in the search space. As a general rule, this occurs during the last part of the braking just before coming to a standstill and leads to critical situations. In order to constantly guarantee a valid target point in the search space of the optimization, the raster points of the search space are varied adaptively, e.g., by a particle swarm. In combination with the filtering of the target vehicle data presented, this increases the robustness of the planning against changing target conditions and, in addition, improves the stopping behavior.

[0062] Furthermore, the stopping process is particularly difficult to adjust or regulate due to the lack of a possibility of “dipping in”. Control deviations which occur (deviation between the planned and actual driven trajectory) cannot be easily corrected, as this would frequently require sections with negative speed (reversing). In order to compensate for the control deviations that occur, various extensions are conceivable and can improve the stopping behavior individually or in combination such as, e.g., by adapting the target conditions of the trajectory planning at low speeds and/or by applying a plateau of a constant acceleration and/or by blending the trajectory planning at very low speeds with a virtual bumper.

[0063] When an acceleration plateau is applied, the planning takes account of the fact that a defined intermediate condition is taken up before the final standstill. This guarantees safe moving off by representing a kind of buffer zone in which any control deviations which exist can be compensated particularly with respect to the distance from the target vehicle. The accurate conversion from the plateau to the standstill can be carried out, e.g., via a pre-controlled acceleration profile. An additional advantage of this approach is the possibility of applying a specific stopping behavior separately from the general trajectory planning.

[0064] A virtual bumper represents a spring-damper system which is virtually affixed or arranged between the vehicle (x.sub.ego) and the target vehicle (x.sub.t) (as depicted in FIG. 8). The result of designing this in a suitable manner is that the vehicle brakes down to a standstill at a specified clearance with respect to the target vehicle, follows at low speeds and can also move off behind the target vehicle. Thanks to the previous or preceding filtering of the nominal clearance, e.g., by means of a spring-damper system, suitable conditions can be created for the handover to the virtual bumper. The virtual bumper additionally offers a safe fallback level in the event that the primary planning does not find a trajectory in the solution range.

[0065] In summary, the modular construction of the coordination layer makes it possible to parameterize and apply individual functionalities separately utilizing the same planner architecture and, thus, makes it possible for the system to be extended by future functionalities in a simple manner. Building on the prior art, the concept of the three-part trajectory is renewed or extended: whilst, e.g., the edges of the acceleration trajectories have previously been fixed, they are now always chosen by a subordinate optimization in a way which is suitable for any initial and final conditions of the trajectory. Further, the trajectory planning is divided into two layers or levels, a coordination or parameterization layer and a planning layer in order to further increase the applicability in serial applications. Whilst the planning layer actually calculates trajectories for traveling on a clear section of road and following vehicles and switches over between these operating modes, the parameterization layer makes it possible to adjust the trajectory properties, e.g., by gain scheduling of the optimization parameters or the like, depending on the situation. The high relevance of the parameterization layer becomes particularly evident when considering human driving profiles. It is true that the optimization supplies trajectories which are optimal within the meaning of the quality measure, the course thereof can however sometimes be unfamiliar to a human driver in some situations, particularly because there is a discrepancy between the mathematical optimum of the trajectory courses and what is perceived to be the optimum by humans. For example, “full speed range ACC” can be provided, which is configured in such a way that a switchover is carried out between the trajectory planner according to the invention and another controller (e.g., the concept of the virtual bumper) in order to control the vehicle in an optimum manner, e.g., including at very low speeds (for example during stopping or crawling). Good dampening control behavior is important especially in a low-speed range in order to ensure, e.g., stability with an optimization-based ACC control concept in a convoy. Further, extensions of the basic functionality of the respective assistance system (e.g., ACC, EBA, etc.) can also be provided such as, e.g., preventing “overtaking on the right maneuvers” or braking before curves. The method according to the invention can further be applied independently of the control structure of the respective means of transportation and consequently offers the possibility of considering the transverse movement of the means of transportation, e.g., an ACC system can consequently serve as a starting point for automated or autonomous driving.

LIST OF REFERENCE NUMERALS

[0066] T1 One-part trajectory (according to the prior art) [0067] T2 Three-part trajectory (according to the prior art) [0068] T3 Three-part trajectory [0069] T4 One-part trajectory (according to the prior art) [0070] T5 Three-part trajectory [0071] T6 One-part trajectory (according to the prior art) [0072] T7 Three-part trajectory [0073] 1 Coordination level [0074] 2 Planning level [0075] 3 Trajectory selection module [0076] 4 Speed module [0077] 5 Distance assistance module [0078] 6 Speed control module [0079] 7 Speed limit assistance module [0080] 8 Cornering assistance module [0081] 9 Speed planner [0082] 10 Distance planner