Trajectory planning system for vehicle and vehicle

Abstract

The invention relates to a trajectory planning system (2) for a vehicle (1), comprising a sensor device (3) for measuring a respective current corresponding actual yaw rate value of a current actual steering angle and an actual curvature of an actual trajectory, and a storage unit (7) in which a target trajectory having a respective target curvature is stored, the trajectory planning system (2) is designed to determine a selected suitable self-steering section (4) and to determine a current self-steering gradient (KEG) in each case by means of a measured actual yaw rate value and a measured actual steering angle located in the respective selected self-steering section (4), and wherein a parameter estimator is provided which estimates the current self-steering gradient (KEG) by means of the measured actual yaw rate value and the measured actual steering angle located in the respective selected self-steering section (4). The present self-steering gradient estimation device is configured to recursively estimate the present self-steering gradient (KEG, tgt) by using the corresponding present self-steering gradient (KEG) as an input. The invention also relates to a vehicle and a method.

Description

Technical Field

[0001] The present invention relates to a trajectory planning system for a vehicle, comprising: a sensor device for measuring a current actual steering angle and a current corresponding actual yaw rate value of an actual curvature of an actual trajectory; and a storage unit in which a target trajectory with a corresponding target curvature is stored. Furthermore, the present invention relates to a vehicle and a method for operating the trajectory planning system. Background Art

[0002] In automated or assisted driving, vehicle trajectory control places high demands on tracking accuracy, stability, and robustness. These characteristics must be maintained throughout the entire trajectory control operating range. The steering system, as the primary interface between trajectory control and the vehicle's lateral dynamics, plays a crucial role in this process. The electronically controlled steering axis follows a predefined guidance path (target displacement trajectory) by retrieving and executing the necessary steering angle setting (target curvature) from memory based on the current vehicle position.

[0003] However, this is not an easy task, as the transfer chain from target curvature to actual curvature is subject to many variable or unknown influences, such as road contact, vehicle load, tires, etc.

[0004] Due to the above effects, choosing the appropriate steering angle preset to produce the desired actual curvature is a significant challenge.

[0005] Inaccuracies arising during the follow-up control of the target curvature calculated within the framework of trajectory control can severely limit the overall system performance and make it difficult to achieve high accuracy requirements.

[0006] Document EP 2977297 B1 discloses a method for determining the current steering angle of a motor vehicle, wherein a steering wheel angle given by a steering device of the motor vehicle is detected by means of a sensor device of the motor vehicle, and the steering angle set due to a presetting by the steering device is determined by means of a control device of the motor vehicle using a predetermined steering angle characteristic curve that describes the relationship between the steering angle and the steering wheel angle, wherein a delay function that describes the time delay between the presetting of the steering wheel angle and the setting of the steering angle is determined, and the steering angle is additionally determined by the control device as a function of the delay function. Summary of the Invention

[0007] The object of the present invention is to provide an improved trajectory planning system for a vehicle, a vehicle itself, and a method for operating such a trajectory planning system.

[0008] This object is achieved by a trajectory planning system having the features of claim 1 , a vehicle having the features of claim 14 , and a method having the features of claim 15 .

[0009] Further advantageous measures are listed in the dependent claims, which can be combined with one another as appropriate to achieve further advantages.

[0010] This object is achieved by a trajectory planning system for a vehicle, the trajectory planning system comprising a sensor device for measuring a current actual steering angle and a corresponding current actual yaw rate value of an actual curvature of an actual trajectory, and a storage unit in which a target trajectory with a corresponding target curvature is stored, wherein the trajectory planning system is configured to determine a selected suitable self-steering section and to determine a current self-steering gradient correspondingly with the measured actual yaw rate value and the measured actual steering angle in the corresponding selected self-steering section, and wherein a parameter estimator is provided, which is configured to recursively estimate a current self-steering gradient with the corresponding current self-steering gradient as input.

[0011] The self-steering gradient is a vehicle-specific factor that can be used to characterize the driving behavior of a motor vehicle. It indicates, for example, whether the steering angle needs to be increased or decreased in order to be able to drive with the same turning radius when the lateral acceleration increases, i.e., the speed in the bend increases.

[0012] The current self-steering gradient is a current self-steering gradient corresponding to a corresponding actual yaw rate value and an actual steering angle; the current self-steering gradient corresponds, for example, to a self-steering gradient calculated over a period of time.

[0013] By using the trajectory planning system according to the present invention, a quasi-stationary transfer characteristic between the target curvature and the actual curvature to be set can be approximated by estimating the current self-steering gradient.

[0014] Here, a quasi-stationary process can be viewed as a series of equilibrium states.

[0015] According to the present invention, it has been recognized that by estimating a quasi-stationary transfer characteristic between the target curvature and the actual curvature, it is possible to reduce deviations in the control of the target curvature. The trajectory planning system according to the present invention makes it possible to estimate the quasi-stationary transfer characteristic between the target curvature and the actual curvature to be set using the current self-steering gradient, even if the quasi-stationary transfer characteristic has a highly nonlinear variability and depends on various vehicle conditions, such as vehicle speed, lateral acceleration, longitudinal acceleration, and tire-road contact.

[0016] Therefore, the current self-steering gradient is used to determine the quasi-stationary transfer characteristics in multidimensional space.

[0017] The trajectory planning system according to the present invention allows the steering angle presetting to be selected so that the required actual curvature is set using the target steering angle, thereby avoiding inaccuracies that occur when following the target curvature calculated by the trajectory control.

[0018] This improves the overall system performance and allows the high precision requirements, for example, for autonomous vehicles to be met.

[0019] The trajectory planning system according to the present invention can recursively determine a multidimensional self-steering gradient surface of the current self-steering gradient using the current self-steering gradient. The driving state associated with the self-steering gradient can be integrated into the parameter estimator in a suitable form.

[0020] In another embodiment, the trajectory planning system is designed to form the current self-steering gradient from the following difference:

[0021]

[0022] Among them, K EG is the current self-steering gradient, δ front,mess is the measured steering angle, l v is the wheelbase, and is the actual yaw rate value measured.

[0023] This makes it easy to determine the current self-steering gradient.

[0024] Furthermore, in another embodiment, the trajectory planning system can be configured to consider selecting a suitable self-steering segment only when the vehicle speed is greater than a predetermined first limit, the absolute value of the actual yaw rate value is greater than a predetermined second limit, and the absolute value of the steering angle change is less than a predetermined third limit for a predetermined time period. In other words, this self-steering segment is only suitable when the vehicle speed is greater than the predetermined first limit, specifically, a vehicle speed determined in this application, for example, greater than 3 km / h. Furthermore, the absolute value of the actual yaw rate value in the selected self-steering segment must be greater than a second limit; in particular, a second limit of 5 degrees / second is suitable. Furthermore, the absolute value of the steering angle change, which can be calculated, for example, as a derivative of the measured actual steering angle, is preferably less than a third limit for a predetermined time period. A third limit of 3 degrees / second and a time period of 2 seconds are suitable, for example. Therefore, in a preferred embodiment, the current self-steering gradient is determined only in suitable segments, such as self-steering segments in which the vehicle state does not change while in a curve.

[0025] Furthermore, the parameter estimator is designed to determine a new current self-steering gradient only when a new current self-steering gradient occurs, thereby saving computing power.

[0026] Furthermore, the previously determined current self-steering gradient can always be used to determine the steering angle input; therefore, real-time determination is not necessary. This prevents overburdening the vehicle's computing power. This is particularly advantageous for autonomous or partially autonomous vehicles.

[0027] Furthermore, in another embodiment, the parameter estimator may be configured to recursively estimate the current self-steering gradient using the current self-steering gradient and current vehicle parameters as inputs, wherein the vehicle parameters include at least the vehicle speed.

[0028] In another embodiment, the parameter estimator may be configured to recursively estimate the current self-steering gradient using the current self-steering gradient, current vehicle parameters, and current environmental data as inputs, wherein the vehicle parameters include at least the vehicle speed.

[0029] These vehicle parameters / environmental data can be detected by means of suitable sensor systems, such as a tachometer, or can be predetermined, such as the vehicle type.

[0030] This allows relevant influencing factors such as vehicle speed, tire values, road surface conditions, longitudinal acceleration, lateral acceleration, tire-road contact, etc., as well as other environmental / vehicle data to be taken into account. The parameter estimator then determines the current self-steering gradient using current / previous results and current environmental / vehicle data.

[0031] This means that the parameter estimator can determine a multi-dimensional self-steering gradient surface based on vehicle speed, longitudinal acceleration, lateral acceleration, tire-road contact characteristics, etc. This allows the quasi-stable transfer characteristics of the vehicle to be determined based on various influencing factors.

[0032] In another embodiment, the parameter estimator can be designed as a recursive least squares estimator. In particular, the recursive least squares estimator can have a forgetting factor to forget older self-steering gradients, that is, older results.

[0033] This allows for stable and rapid determination of the current self-steering gradient. Recursiveness allows for the online use of currently generated data while maintaining the same complexity in each recursive step. In particular, a forgetting factor can be introduced that forgets older results, i.e., older self-steering gradients. Consequently, the importance of historical data for optimization is reduced, while current data can be given greater weight. Alternatively, the parameter estimator can be constructed, for example, as an artificial neural network with nodes to be determined.

[0034] Furthermore, in another embodiment, the parameter estimator has internal parameters to be optimized for determining the current self-steering gradient as output. When the parameter estimator is configured as an artificial neural network, these internal parameters may be, for example, nodes. In this case, the parameter estimator is preferably configured to use these parameters for future estimations of the current self-steering gradient as output once the internal parameters to be optimized, such as nodes, have reached a certain accuracy value.

[0035] This means that, starting from a certain quality, the parameter estimator only serves as a known optimization function in space, into which the input data are entered and with the help of which the current self-steering gradient is determined as output, for example using already optimized nodes (if the parameter estimator is an artificial neural network).

[0036] Furthermore, in another embodiment, the trajectory planning system may be configured to form a preset angle using the current self-steering gradient so that the target curvature can be converted into the desired actual curvature in a timely manner.

[0037] In addition, the trajectory planning system can be configured to determine the preset angle using the following formula:

[0038]

[0039] Among them, δ ff is the pre-adjustment angle, l v is the current wheelbase, v CR is the vehicle speed, K EG,tgt is the current self-steering gradient, and κ CR,tgt is the target curvature.

[0040] Furthermore, in a further embodiment, a steering system is provided which has a steering parameter for setting a steering angle, the steering system being designed to continuously set the steering parameter by means of a preset angle.

[0041] In this case, the target steering angle sent as a request to the steering system can consist of a pre-set steering angle and a feedback steering angle. In this case, in a quasi-stationary state without disturbances, if the self-steering gradient is known, the pre-set steering angle preferably directly leads to the desired vehicle behavior, and the feedback steering angle is zero.

[0042] Furthermore, the object is also achieved by a vehicle having a trajectory planning system as described above, wherein the vehicle has a receiving unit and / or a trajectory generation system, the receiving unit being configured to receive a target trajectory with a target curvature to be set at a predetermined position by one or more vehicles traveling in front, and the trajectory generation system being configured to generate the target trajectory with the aid of at least navigation data and environmental data.

[0043] Thus, the vehicle can be, for example, a trailing vehicle following a leading vehicle, or it can be an autonomous vehicle that uses navigation data and environmental data recorded by a sensor system to create a target trajectory. The trajectory planning system according to the present invention requires minimal computing power and, in particular, does not impose an additional burden on computing power for autonomous vehicles, which require a lot of computing power.

[0044] Furthermore, the object is achieved by a method for operating a trajectory planning system for a vehicle, the method comprising the following steps:

[0045] - measuring the current actual steering angle and the corresponding current actual yaw rate value of the actual curvature of the actual trajectory;

[0046] -providing a target trajectory with a corresponding target curvature;

[0047] - Determine the appropriate self-steering section selected by the trajectory planning system;

[0048] - determining the respective current self-steering gradient using the measured actual yaw rate value and the measured actual steering angle in the respectively selected suitable self-steering segment;

[0049] - Input the corresponding current self-steering gradient into the parameter estimator, and recursively estimate the current self-steering gradient with the help of the parameter estimator.

[0050] In this case, the trajectory planning system can in particular be the trajectory planning system described above. Furthermore, the advantages of the trajectory planning system can also be transferred to the method.

[0051] Other features and advantages of the present invention will be apparent from the following description made with reference to the accompanying drawings. A person skilled in the art may derive variations thereof without departing from the scope of protection of the present invention as defined by the following claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings schematically show:

[0053] Figure 1 A vehicle having a trajectory planning system according to the present invention is shown;

[0054] Figure 2 The determination of the self-steering section is shown;

[0055] Figure 3 Shows the current self-steering gradient K EG and the current self-steering gradient K EG,tgt ;

[0056] Figure 4 Shows the current self-steering gradient KEG and the current self-steering gradient K EG,tgt ;

[0057] Figure 5 An overview of a trajectory planning system and method thereof according to the present invention is shown. DETAILED DESCRIPTION

[0058] Figure 1 A vehicle 1 with a trajectory planning system 2 according to the present invention is shown schematically.

[0059] If vehicle 1 is designed as an autonomous vehicle, a trajectory generation system (not shown) may be provided for generating a target trajectory from navigation data and environmental data, such as surrounding data such as other road users and stationary objects. The target trajectory and its target curvature are stored in storage unit 7.

[0060] Furthermore, the trajectory planning system 2 includes a sensor device 3. The sensor device 3 is configured to detect the current actual yaw rate value of the actual curvature and the corresponding current actual steering angle. To this end, the sensor device 3 may include a rotation angle sensor and other sensors for detecting the actual yaw rate value and the actual steering angle.

[0061] Furthermore, a steering system 5 is provided, which has a steering manipulated variable and is used to convert the target curvature into an actual curvature using the steering manipulated variable. To this end, the steering system 5 can include actuators, such as rotary actuators, and sensors. Vehicle 1 can also be configured as a trailing vehicle. Such a vehicle receives the target yaw rate value to be set from a preceding vehicle traveling along the same target trajectory. For this purpose, vehicle 1 can include a corresponding receiving unit (not shown), for example, for receiving the target curvature via radio (vehicle-to-vehicle connection).

[0062] Trajectory control, especially for setting the curvature of a vehicle 1 in the context of automated or assisted driving, places high demands on tracking accuracy, stability, and robustness. These characteristics must be maintained across the entire operating range of trajectory control. A key challenge in trajectory control is accurately following the target curvature.

[0063] In this trajectory control framework, the target steering angle is calculated from the target curvature using the vehicle kinematic model and the self-steering gradient as follows:

[0064] δ front =δ ff +δ fb (1)

[0065] Here, the target steering angle δ sent as a request to the steering system 5 is front By pre-adjusting the steering angle δ ffand feedback steering angle δ fb composition.

[0066] In order to determine the required pre-adjusted steering angle δ ff , the trajectory planning system 2 is based on the actual yaw rate value measured and the measured steering angle δ mess Determine the current self-steering gradient:

[0067]

[0068] Among them, l v is the current wheelbase, and the curvature of the vehicle in the quasi-stationary state is obtained from the yaw rate through the following formula:

[0069]

[0070] The current self-steering gradient is only calculated if a suitable self-steering segment 4 is present. A self-steering segment 4 is always present in the following cases:

[0071] The vehicle speed is greater than a predetermined first limit value. This first limit value is preferably greater than 3 km / h.

[0072] - The absolute value of the current actual yaw rate value is greater than a predetermined second limit value. To this end, the second limit value can be set to 5 degrees / second, for example.

[0073] The absolute value of the change in the steering angle is less than a predetermined third limit value for at least a predetermined time period. In this case, the change in the steering angle can be determined as a derivative of the measured actual steering angle. The third limit value can be set, for example, to 3 degrees / second, and the time period can be set, for example, to 2 seconds.

[0074] In general, this means that, for example, when negotiating a curve, the vehicle state must not change within a predetermined time limit.

[0075] If there is a self-steering section 4, the following formula can be used:

[0076]

[0077] Determine the current self-steering gradient K in the corresponding self-steering section 4 EG ,in Equal to the actual yaw rate value measured, δ mess is equal to the measured steering angle, and l v Equal to the current wheelbase.

[0078] Figure 2 The diagram shows the process of determining the self-steering section 4 in which such calculations can be performed.

[0079] Here, the first upper graph shows the change in the current vehicle speed over time.

[0080] The second graph below it shows the measured current actual yaw rate value as a function of time.

[0081] The third graph shows the measured change in the current actual steering angle over time. The fourth graph shows the self-steering segment 4 that is formed, in which the vehicle speed is greater than a predetermined first limit value, the absolute value of the current actual yaw rate value is greater than a predetermined second limit value, and the absolute value of the steering angle change is less than a predetermined third limit value for at least one predetermined time period.

[0082] The trajectory planning system 2 also has a parameter estimator, which is particularly designed as a recursive least squares estimator 6. Such an estimator can provide results quickly and reliably. The recursive nature allows the use of currently generated data online while maintaining the same complexity in each recursive step. Alternatively, the parameter estimator can also be designed as an artificial neural network, for example.

[0083] Here, the recursive least squares estimator 6 obtains the current self-steering gradient K of the current self-steering section 4 EG Here, the recursive least squares estimator 6 has internal parameters to be optimized to determine the current self-steering gradient K EG,tgt as output.

[0084] Furthermore, the recursive least squares estimator 6 receives as input vehicle parameters, in particular vehicle speed and environmental data such as road conditions and, for example, friction coefficient, longitudinal acceleration, lateral acceleration, tire-road contact.

[0085] With the help of vehicle parameters and environmental data as well as the current self-steering gradient K EG , the current self-steering gradient K is now recursively determined by the recursive least squares estimator 6 EG,tgt .

[0086] This current self-steering gradient can be determined online with reasonable computation and storage requirements.

[0087] Here, a forgetting factor can be introduced, which forgets results that are too old.

[0088] Likewise, the recursive least squares estimator 6 can be constructed such that once the internal parameters to be optimized reach a certain accuracy value, these parameters are used to estimate the current self-steering gradient K in the future. EG,tgt as output.

[0089] This means that, starting from a certain quality, the recursive least squares estimator 6 only uses a known function in space, into which the input data are input and with the aid of which the current self-steering gradient is determined as output.

[0090] Only when there is a newly determined current self-steering gradient K EG and the new current self-steering section 4, the trajectory planning system 2 determines the new current self-steering gradient K by means of the recursive least squares estimator 6 EG,tgt Since the previous current self-steering gradient K is used before EG,tgt , so there is no need to consume excessive computing power for real-time calculations. This reduces the computational burden on the vehicle. This is particularly beneficial for autonomous or partially autonomous vehicles.

[0091] The trajectory planning system 2 can then use the current self-steering gradient K determined thereby to EG,tgt To determine the pre-adjusted steering angle δ ff , the pre-adjusted steering angle here sets the target curvature κ CR,tgt Converted to target steering angle δ front :

[0092]

[0093] Among them, l v represents the wheelbase, and v CR Indicates vehicle speed.

[0094] Here, the target steering angle δ is sent as a request to the steering system 5. front By pre-adjusting the steering angle δ ff and feedback steering angle δ fb composition:

[0095] δ front =δ ff +δ fb (1)

[0096] In the absence of interference and quasi-stationary conditions, if the self-steering gradient is known, then the pre-adjusted steering angle δ ff directly leads to the desired vehicle behavior and feedback steering angle δ fb is zero.

[0097] Then, the current self-steering gradient K EG,tgt The current driving state is continuously used for the target curvature control, so that the current self-steering gradient K estimated for the current driving state is always taken into account in the pre-control. EG,tgt This means that the current self-steering gradient K EG,tgt Continuation is used in curvature control, ie, in Equation 3 and Equation 1.

[0098] If the target curvature CR,tgt is zero, then the pre-adjusted steering angle δ in equation 3 is ff Also zero.

[0099] By means of the trajectory planning system 2 according to the invention and the current self-steering gradient K determined according to the invention EG,tgt , enabling online estimation of the quasi-stationary transfer characteristic and thereby reducing the deviation between the target curvature and the actual curvature. The trajectory planning system 2 according to the present invention takes into account the strong nonlinearity of the vehicle's quasi-stationary transfer characteristic and its relationship to various vehicle conditions, such as vehicle speed, lateral acceleration, longitudinal acceleration, and tire-road contact.

[0100] The trajectory planning system 2 eliminates inaccuracies that occur when following the target curvature calculated by the trajectory control, thereby improving the overall system performance and meeting high accuracy requirements, particularly for vehicles operating autonomously or following behind.

[0101] Figure 3 The current self-steering gradient K determined is shown EG and the current self-steering gradient K EG,tgt The relationship between the vehicle speed and the vehicle speed. Here, the first curve diagram above shows the current self-steering gradient K EG (act ssg immediate) and the current self-steering gradient K EG,tgt (act ssg estimated) and changes over time.

[0102] Furthermore, the second middle graph shows temporal changes in a self-steering section 4 formed, in which the vehicle speed is greater than a predetermined first limit value, the absolute value of the current actual yaw rate value is greater than a predetermined second limit value, and the absolute value of the steering angle change is less than a predetermined third limit value for at least a predetermined period of time.

[0103] The third graph below shows the change in vehicle speed over time.

[0104] Figure 4 The current self-steering gradient K determined is shown EG and the current self-steering gradient K EG,tgt Relationship with vehicle speed, actual yaw rate value and target yaw rate value.

[0105] Here, the first graph above shows the current self-steering gradient K EG (act ssg immediate) and the current self-steering gradient K EG,tgt (act ssg estimated) and changes over time.

[0106] In addition, the second graph in the middle shows the change of the actual yaw rate value (act yaw rate) and the target yaw rate value (tgt yaw rate) over time. EG,tgt , the target yaw rate value (tgtyaw rate) and the actual yaw rate value (act yaw rate) become consistent, and thus the target curvature and the actual curvature are kept consistent in the quasi-stationary driving state.

[0107] The third graph below shows the change in vehicle speed over time.

[0108] Figure 5 An overview of a trajectory planning system 2 and its method is shown.

[0109] In a first step, the target curvature to be set is stored in the memory unit 7. The corresponding current actual yaw rate value and the corresponding current actual steering angle are also detected. The vehicle speed and the steering angle change are then determined.

[0110] Furthermore, a self-steering section 4 is determined in which the vehicle speed is greater than a first limit value, the absolute value of the current actual yaw rate value is greater than a second limit value, and the absolute value of the steering angle change is less than a third limit value for at least a predetermined period of time.

[0111] The actual yaw rate values ​​and the corresponding actual steering angles in the self-steering segments 4 are used to determine the current self-steering gradient K EG :

[0112]

[0113] in, Equal to the actual yaw rate value measured, δ mess is equal to the measured steering angle, and l v Equal to the current wheelbase.

[0114] Then, the current self-steering gradient K EG and at least vehicle speed, friction coefficient / road condition, longitudinal acceleration, lateral acceleration, and tire-road contact are input as parameters into a recursive least squares estimator 6 which estimates the current self-steering gradient K based on at least some previous results. EG,tgt .

[0115] With the help of the current self-steering gradient K EG,tgt , determine the pre-adjustment angle, which is input into the steering system 5 for curvature control to compensate for the deviation in converting the target curvature into the actual curvature to be set.

[0116] List of reference numerals:

[0117] 1 vehicle

[0118] 2 Trajectory Planning System

[0119] 3 Sensor device

[0120] 4 Self-steering section

[0121] 5 Steering system

[0122] 6 Recursive Least Squares Estimator

[0123] 7 storage units.

Claims

1. A trajectory planning system (2) for a vehicle (1), the trajectory planning system comprising: A sensor device (3) and a storage unit (7), wherein the sensor device is used to measure the current actual steering angle and the actual curvature of the actual trajectory, and the corresponding current actual yaw rate value, wherein the storage unit stores a target trajectory with a corresponding target curvature. It is characterized in that The trajectory planning system (2) is configured to determine a selected suitable self-steering segment (4) and to determine a current self-steering gradient (K) correspondingly using the measured actual yaw rate value and the measured actual steering angle in the corresponding selected self-steering segment (4). EG ), and wherein a parameter estimator is provided, which is configured to use the corresponding current self-steering gradient (K EG ) as input, recursively estimate the current self-steering gradient (K EG,tgt ).

2. The trajectory planning system (2) according to claim 1, characterized in that The trajectory planning system (2) is configured to form the current self-steering gradient (K EG ): Among them, (δ front,mess ) is the measured steering angle, (l v ) is the wheelbase, and is the actual yaw rate value measured.

3. The trajectory planning system (2) according to any one of the preceding claims, characterized in that The trajectory planning system (2) is configured to consider selecting a suitable self-steering section (4) only when the vehicle speed is greater than a predetermined first limit value, the absolute value of the actual yaw rate value is greater than a predetermined second limit value, and the absolute value of the steering angle change is less than a predetermined third limit value within a predetermined time period.

4. The trajectory planning system (2) according to claim 3, characterized in that The parameter estimator is constructed to be effective only when there is a new current self-steering gradient (K EG ) is determined only when the new current self-steering gradient (K EG,tgt ).

5. Trajectory planning system (2) according to any one of the preceding claims, characterized in that The parameter estimator is constructed by using the current self-steering gradient (K EG ) and the current vehicle parameters as input, recursively estimate the current self-steering gradient (K EG,tgt ), wherein the vehicle parameters include at least the vehicle speed.

6. The trajectory planning system (2) according to any one of the preceding claims 1 to 4, characterized in that The parameter estimator is constructed by using the current self-steering gradient (K EG ) and the current vehicle parameters and current environment data as input, recursively estimate the current self-steering gradient (K EG,tgt ), wherein the vehicle parameters include at least the vehicle speed.

7. The trajectory planning system (2) according to claim 6, characterized in that The environmental data includes at least road conditions and / or a detected friction coefficient.

8. Trajectory planning system (2) according to any one of the preceding claims, characterized in that The parameter estimator is constructed as a recursive least squares estimator (6).

9. The trajectory planning system (2) according to claim 8, characterized in that The recursive least squares estimator (6) has a forgetting factor for forgetting the result of the recursive least squares estimator (6) from a predetermined time point.

10. Trajectory planning system (2) according to any one of the preceding claims, characterized in that The parameter estimator has internal parameters to be determined to determine the current self-steering gradient (K EG,tgt ) as output, wherein the parameter estimator is constructed to use the parameter for future estimation of the current self-steering gradient (K once the internal parameter to be determined reaches a certain accuracy value EG,tgt ) as output.

11. Trajectory planning system (2) according to any one of the preceding claims, characterized in that The trajectory planning system (2) is configured to use the current self-steering gradient (K EG,tgt ) forms a pre-adjustment angle (δ ff ), so that the target curvature can be converted into the desired actual curvature in a timely manner.

12. The trajectory planning system (2) according to claim 11, characterized in that The trajectory planning system (2) is configured to determine the pre-adjustment angle (δ ff ): Among them, (δ ff ) is the pre-adjustment angle, (l v ) is the current wheelbase, (v CR ) is the vehicle speed, (K EG,tgt ) is the current self-steering gradient, and (κ CR,tgt ) is the target curvature.

13. The trajectory planning system (2) according to claim 11 or 12, characterized in that The trajectory planning system (2) comprises a steering system (5) having a steering parameter for setting a steering angle, wherein the steering system is configured to ff ) continuously sets the steering parameters.

14. Vehicle (1) having a trajectory planning system (2) according to any one of the preceding claims, wherein: The vehicle (1) has a receiving unit and / or a trajectory generation system, wherein the receiving unit is used to receive a target trajectory with a target curvature to be set at a predetermined position by one or more vehicles traveling in front, and the trajectory generation system is used to generate the target trajectory with the help of at least navigation data and environmental data.

15. A method for operating a trajectory planning system (2) for a vehicle (1), the method comprising the following steps: - measuring the current actual steering angle and the corresponding current actual yaw rate value of the actual curvature of the actual trajectory; -providing a target trajectory with a corresponding target curvature; - determining a selected suitable self-steering section (4) by the trajectory planning system (2); - Determine the corresponding current self-steering gradient (K) using the measured actual yaw rate value and the measured actual steering angle in the corresponding selected suitable self-steering section (4) EG ); - The corresponding current self-steering gradient (K EG ) is input into the parameter estimator, and the current self-steering gradient (K EG,tgt ).

Images