Method, control means, vehicle, and computer program for implementing lane guidance in a vehicle

Abstract

Method, control means, vehicle, and computer program for implementing lane guidance in a vehicle The present invention relates to a method for implementing lane guidance of a vehicle (1), in which a control variable is determined via a lateral feedback control (7) based on a lateral spacing error describing the lateral deviation (4) of the vehicle (1) relative to a predetermined target trajectory (5), and the lateral position of the vehicle (1) is adjusted to the target trajectory (5) depending on the control variable, wherein at least one calibration value is additionally applied to the control variable, the calibration value being determined by at least one calibration function (12) at least partly depending on the time integral of the lateral spacing error.

Description

[Technical Field]

[0001] The present invention relates to a method for implementing lane guidance, in which a control variable is determined via lateral feedback control based on a lateral spacing error describing the lateral deviation of the vehicle from a predetermined target trajectory, and the lateral position of the vehicle is adjusted on the target trajectory in response to the control variable. The present invention also relates to a control means, a vehicle, and a computer program. [Background technology]

[0002] Vehicles can be equipped with systems or functions for automatic lane guidance, which automatically adjust the lateral position of the vehicle while it is moving to a predetermined trajectory. The lateral position of the vehicle, i.e., the lateral position of the vehicle within the driving lane, can be changed or adjusted to a target position, for example, by automated steering operations.

[0003] Such lane-keeping functions typically use sensors, such as cameras, radar or lidar systems, for capturing the vehicle's surroundings, and / or surrounding data obtained by a navigation system, such as a GPS, to guide the vehicle along a planned trajectory. The trajectory is set to minimize strong or abrupt changes in the vehicle's lateral position, for example, to ensure a high level of driving comfort. Guiding the vehicle along such a trajectory is typically implemented as a feedback control that continuously determines at least the vehicle's lateral deviation from the planned trajectory, and outputs control commands of appropriate magnitude and direction to the vehicle's, for example, electric steering system, depending on the deviation. Criteria for determining the quality of the feedback control of such systems include, for example, the amplitude and dynamic characteristics of the lateral feedback deviation and the frequency of jerky movements of the vehicle's steering wheel. Ideally, the lateral feedback error is zero in all driving situations, and lane-guiding steering interventions or resulting steering movements are solely due to changes in the planned trajectory.

[0004] The performance of such lane guidance depends on the quality and performance of the sensors and actuators used in the vehicle. In principle, the use of multiple sensors or actuators and / or the implementation of appropriate software measures can improve the overall system performance to some extent. This also applies in particular to driver assistance systems such as functions to support lane keeping.

[0005] However, sensor sensing errors, i.e., any deviation between the sensor signal and its physical counterpart, not only affect the character of the feedback control, but also the quality of the feedback control of the vehicle's lane guidance, especially due to the inheritance of these errors.To counter this, partly different but partly overlapping action chains are applied depending on the sensor information.

[0006] Image distortions of a camera capturing lane markings, for example, can cause the calculated curvature of the road to deviate from the actual curvature, a so-called ground truth error. Lane guides typically include at least one feedback control and / or predictive control that adjusts the lateral vehicle position according to at least one determined lane curvature. Curvature information describing the lane curvature is often calculated at a predicted point ahead of the vehicle. This information can be used to calculate the steering angle required for each curvature using an inverse vehicle model and input this as a target steering angle component to a steering angle feedback control, for example as part of predictive control.

[0007] If the curvature information has an offset due to an error compared to the actual curvature, the predicted target steering angle also has an offset component that acts like a disturbance on the lateral feedback control. This has the effect that when the lane-keeping assistant is activated, the vehicle does not follow the trajectory as expected, but rather "maintains a continuous lateral deviation from the planned trajectory" or "at least temporarily deviates until the planned target trajectory is followed after the adjustment process has decayed," depending on the implementation of the lateral feedback control or trajectory-following feedback control—i.e., whether a feedback control type with or without considering static accuracy is used. The continuous or temporary lateral deviation then affects driving comfort, which can be a criterion for the driver's acceptance of the driver assistance system.

[0008] The degree of camera image distortion typically fluctuates during the operation of the driver assistance system due to the ongoing execution of camera calibration routines. This means that depending on the state of the camera calibration, the value of the curvature and predictive control offset, and the resulting adverse impact on vehicle lateral guidance, will change over time, even if the camera transfer characteristic and the resulting curvature offset are band-limited or gradual rather than abrupt.

[0009] Because the steering angle is often a significant auxiliary or feedback control variable in lane tracking assistance systems, it is frequently used in routines for calibrating the steering angle offset. The calculated steering angle offset always contains uncompensated components due to the finite accuracy of the calibration. Residual offset errors in the steering angle signal can have several possible causes. Typical calibration routines are performed using signals from yaw rate sensors or wheel rotation speed sensor systems. Errors in these sensors, such as offset and / or linearity errors, and in the case of wheel rotation speed sensors, differences in rotation speed due to differences in tire pressure on one axle, are also inherited in the calculated steering angle offset. The effect of the steering angle offset is therefore similar to that of the curvature offset described above: Even in this case, when the lane keeping assistant is activated, the vehicle does not follow the predetermined trajectory as expected, but has permanent or temporary lateral deviations from the planned trajectory, depending on the implementation of the lateral feedback control or trajectory following feedback control, which negatively affects comfort and acceptance of the driver assistance system. This is equally true for uncompensated front and rear steering angle offsets.

[0010] The sensor signals for yaw rate, front and rear axle steering angles, and lateral acceleration can be used for other vehicle functions that can affect the lateral position or lateral guidance of the vehicle. For example, these sensor signals can be used to estimate disturbance forces and moments acting on the vehicle, as well as the vehicle's sideslip angle due to, for example, the influence of road inclination or crosswinds. In this case, the calculated values ​​also contribute to the target steering angle for lateral guidance of the vehicle as part of the disturbance value interruption. However, the corresponding offset of the sensor signals, in particular, is also passed on as an error to the target steering angle and affects the lateral position of the vehicle. As a result, here too, undesirable temporary or continuous lateral deviations occur in the lane-guided vehicle.

[0011] DE 10 2008 026 233 B4 discloses a method for compensating for a steering angle offset of a motor vehicle using a vehicle model. A yaw rate is calculated based on the acquired steering angle, the vehicle speed, and an assumed steering angle offset. The calculated yaw rate is then compared with a measured yaw rate, and the difference between them forms an error equation for iteratively determining the steering angle offset, which is then fed back to the vehicle model for compensation.

[0012] Such yaw rate and lateral acceleration offset calibrations have finite accuracy and will always track the actual offset, so performing this additional calibration may result in unwanted steering intervention. [Prior art documents] [Patent documents]

[0013] [Patent Document 1] DE 10 2008 026 233 B4 Summary of the Invention [Problem to be solved by the invention]

[0014] The object of the present invention is therefore to improve the accuracy of lane guidance in a vehicle, even under the influence of offset errors, and thereby to increase comfort and driver acceptance of the assistance system. [Means for solving the problem]

[0015] To achieve this object, the present invention provides a method of the type mentioned at the beginning, in which at least one calibration value is additionally applied to the controlled variable, the calibration value being determined by at least one calibration function at least partly as a function of the time integral of the lateral distance error.

[0016] That is, the feedback control of the lateral position of the vehicle is performed using the sum of the control variable and at least one calibration value. Depending on the configuration of the lateral feedback control and / or lane guide, other factors may be applied to the control variable in addition to the at least one calibration value.

[0017] The control variable to which the calibration value is applied influences the lateral position of the vehicle. The control variable may be, for example, a target steering angle. For example, this control variable can be set by a steering angle interface connected to an electric power steering (EPS). Alternatively, the control variable can be set in a moment interface that influences the steering torque or steering angle of the EPS.

[0018] The calibration value, which is determined by the calibration function depending, inter alia, on the sign and amplitude of the incorporated lateral spacing error, is applied to the controlled variable in such a way that the lateral spacing error is always reduced, i.e., the effect of applying the calibration value to the controlled variable is that the target steering angle is corrected or compensated for by the lateral spacing error integrated over time.

[0019] Lateral spacing error describes the lateral or sideways displacement of the vehicle, i.e., the deviation of the vehicle from a predetermined target trajectory. Therefore, lateral spacing error is also called lateral offset or lateral deviation. In other words, lateral spacing error indicates the deviation of the current measured vehicle position from the desired vehicle target position described by the target trajectory. The determination of the relative lateral deviation of the vehicle with respect to the target trajectory can be referenced to an agreed point along the longitudinal axis of the vehicle, whereby advantageously the lateral deviation can be referenced to the geometric center of the rear or front axle or to any forward predicted point set along the longitudinal axis of the vehicle.

[0020] The lateral spacing error can be determined continuously, for example, based on sensor data from at least one vehicle peripheral sensor. The peripheral sensor can be, for example, a camera, a lidar sensor, a radar sensor, and / or an ultrasonic sensor. Additionally or alternatively, the lateral spacing error can be determined based on position data from a navigation system. The lateral spacing error also has a time-dependent value that is integrated by a calibration function.

[0021] The lateral feedback control means used in the lateral feedback control can be, for example, a tracking feedback control means or can include a similar means. A calibration function, which can also be called a centering calibration or centering function, is then provided in addition to the tracking feedback control and constitutes a parallel or higher-level feedback control loop that influences the controlled variable, i.e., the lateral position of the vehicle. In particular, the dynamic characteristics of the calibration function are significantly smaller than those of the lateral feedback control, so that the lateral spacing error integrated over time only affects the controlled variable very slowly.

[0022] As a basic calculation rule for determining the calibration function or the calibration value, for example, an integration means with a relatively small integration coefficient can be used, which allows integration of the lateral spacing error with a small dynamic characteristic, resulting in a dynamic characteristic that is also small in influence of the calibration value.

[0023] According to the invention, at least one calibration function can be implemented as an I· element or a PT1· element, respectively. The PT1· element has similar integration characteristics compared to a pure I· element, but has the advantage that it does not require measures to limit the integrator state, which is required for an I· element. This advantage must be weighed against the disadvantage that, when utilizing the parameterization freedom provided by the PT1· element, it is not possible to achieve strict compensation for the offset in the transient state in each individual application.

[0024] The adaptive dynamics of the calibration value are set to be in particular comparable to the dynamics of the change in the offset error, e.g., the maximum expected offset drift due to the temperature of the sensors used in the lateral feedback control, converted into the corresponding steering angle plane. This has the advantage that, in the case of a highly dynamic calibration function, no overshoot of the lateral feedback control occurs, which could trigger the intervention of other compensation functions acting on the steering angle, such as road cross slope compensation or crosswind compensation.

[0025] By using at least one calibration function or applying at least one calibration value to a manipulated variable, offset errors that continuously affect lane guidance can advantageously be partially or completely reduced over time or partially or completely compensated for over time. Compensation functions that affect steering angle often do not operate continuously. For example, a lane keeping function typically only becomes effective after the driver activates it. If the compensation function detects an apparent feedback control error due to an offset error during activation, the compensation function outputs a controlled variable in an attempt to reduce the feedback control error. This may result in abrupt steering movements, depending on the dynamic characteristics of the compensation function. Compensating for offset errors in advance advantageously avoids abrupt steering movements to correct accumulated lateral deviations, thereby improving occupant comfort while lane guidance is activated. This also improves acceptance of lane guidance or driver assistance systems that implement lane guidance.

[0026] In summary, the basic concept of the high-level calibration function can be interpreted as a low-dynamic constant feedback control that sets a target value for the lateral spacing error to zero. As a result, during automatic lateral guidance or lane guidance, the sum of the direct and indirect offset effects of all sensors used in the system is reduced with respect to the lateral deviation of the vehicle.

[0027] According to the present invention, the time constant of the calibration function may be configured to be at least twice as large as the dominant time constant of the lateral feedback control. Preferably, the time constant of the calibration function can be set to be at least 5, at least 10, at least 100, or at least 1000 times larger than the dominant time constant of the lateral feedback control. In principle, even larger differences between the time constants are conceivable. For example, a calibration function realized as an I· element can have an integration coefficient in the range of 0.01° / (m*s) to 0.00005° / (m*s).

[0028] In a preferred embodiment of the present invention, an integral value of the calibration function is calculated by integrating the lateral spacing error over time, and this integral value is retained even when the lane guide is stopped, and is used as the initial value of the time integral of the lateral spacing error when the lane guide is restarted.

[0029] The integral value then represents a total value formed by integrating the lateral spacing error over time up to the time point associated with the integral value. In other words, the integral value represents, at a specific time point, the result of integrating the lateral spacing error over time up to that time point. If multiple calibration functions are used, in particular, a separate integral value associated with the calibration function is formed for each calibration function. At least one integral value can be stored in a callable form, for example, in a storage means of a computing means, such as a control device, that executes the method.

[0030] That is, the state of the calibration function, i.e., the state of the integrating means or equivalent functional unit forming the calibration function, is not reset, for example, during the current ignition cycle of the vehicle, but is maintained even if the driver assistance is repeatedly switched on and off or if there is driver intervention. However, it can also be envisaged that the integrating means will be reset if an unreasonable activity is recognized, for example, if a jump is detected or signaled in the output of one or more offset calibrations of the on-board sensors, or if there is a request via the vehicle's diagnostic interface. The integrating value can also be maintained over the ignition cycle of the vehicle, i.e., over several trips.

[0031] In the present invention, it is also possible to envisage that the lateral feedback control includes a predictive control, in which case the predictive control determines a predictive control amount depending on a measured and / or predicted curvature value of the road surface on which the vehicle is traveling, and the control amount includes the predictive control amount as one of its components.

[0032] In this case, the measured curvature values ​​and / or predicted curvature values ​​can be derived, in particular, from sensor data of one or more surrounding sensors of the vehicle. The predictive control, in particular, uses the measured curvature values ​​and / or predicted curvature values ​​to calculate the steering angle required for each curvature via an inverse vehicle model, which can be provided as a predicted control variable, for example, to a control value of a trajectory tracking feedback control means of lateral feedback control. The vehicle model changes its transmission behavior depending on the vehicle speed. This clearly means that an error in the measured curvature values ​​affects the lateral guidance of the vehicle depending on the vehicle speed.

[0033] The present invention envisages implementing predictive control relying on a vehicle model describing the vehicle's inherent steering angle gradient, which is adapted during vehicle operation depending on at least one vehicle parameter, in particular the vehicle weight and / or tire stiffness.

[0034] When a vehicle is cornering, the lateral spacing error resulting from the error in the predictive control depends not only on the curvature offset error but also on the inaccuracy of the stored vehicle parameters, particularly the uncertainty of the specific steering angle gradient used in the vehicle model. The error or deviation in the specific steering angle gradient acts as a disturbance to the vehicle's lateral feedback control and thus partially causes an additional lateral spacing error, which in turn leads to incorrect steering due to the calibration value generated by the calibration function. In such cases, the calibration value responds not only to the sensor offset amount but also to other factors. As long as the error or deviation in the specific steering angle gradient is limited, the dynamic characteristics of the calibration function are low and its impact is minor.

[0035] In order to avoid the influence of errors and / or deviations in the large specific steering angle gradients, these can be adjusted, particularly preferably continuously, as a function of at least one vehicle parameter, for example as a function of the current vehicle mass, the current tire stiffness of one or more tires of the vehicle, and / or other vehicle parameters.

[0036] In a preferred embodiment of the invention, it can be envisaged that the integration over time of the lateral spacing error is carried out using a weighting factor, particularly preferably an integration coefficient that depends on the speed of the vehicle and / or that increases with increasing absolute value of the lateral spacing error.

[0037] For example, while using a constant integration coefficient may be an easily achievable compromise for the entire vehicle speed range, using a speed-dependent integration coefficient or equivalent weighting factor has the advantage of better compensating for speed-dependent offsets. The effect of the offset on the lateral spacing error may be based on, for example, the setting of the feedback control parameters of the normally used lateral feedback control, which depends on the vehicle's driving speed. This changes the disturbance compensation characteristics of the lateral feedback control, and consequently, the effect of the sensor offset on the lateral spacing error. To address this situation, using a speed-dependent weighting factor, such as a speed-dependent integration coefficient, allows the time integration to also depend on the driving speed. Advantageously, this allows for ideal, individually adapted calibration dynamics to be achieved for all driving speed ranges.

[0038] Additionally or alternatively, the weighting factor can depend on the absolute value of the lateral spacing error, i.e., the value of the lateral spacing error to be integrated and / or previously integrated, with larger lateral spacing errors being weighted more strongly. For example, a nonlinear correspondence rule and / or one or more lateral spacing error thresholds can be used to determine an assigned weighting factor or integration coefficient for the current lateral spacing error and to use it in the time integration of the lateral spacing error. By gradually weighting the absolute value of the lateral spacing error, i.e., by increasing the weighting factor or integration coefficient as the absolute value of the lateral spacing error increases, the calibration speed can be effectively improved.

[0039] For example, the integral coefficient and the calibration function adapted thereto can be switched between two or more values, with a larger integral coefficient being selected when the absolute value of the lateral spacing error is larger. Advantageously, simply doubling the integral coefficient or integral constant when the absolute value of the lateral spacing error exceeds 0.2 m allows for an effective and rapid initial adaptation to certain offset situations at the start of the driver assistance function.

[0040] Even in cases where a vehicle has an additional sensor offset calibration function that eliminates the detected offset drift of at least one sensor of the vehicle by a step-wise correction for each offset after a confirmation or damping time has elapsed, a rapid adjustment to the new offset situation can be achieved by increasing the adaptation rate of the centering calibration depending on the absolute value of the lateral spacing error. This becomes more important the longer the damping time of the sensor offset calibration function, because during this period the calibration function performs compensations based on the time integral of the lateral spacing error that must be integrated back after more sensor offset calibration function signal updates. Using large weights or large integration coefficients can advantageously increase the speed of the correction process.

[0041] Since step-like corrections by the vehicle's sensor offset calibration functions cause sudden lateral movements of the lane-guided vehicle depending on the step height, in a vehicle including a control means configured to implement the method of the present invention, all sensor offset calibration functions can be assumed to change their offsets only within a band limit, rather than in a step-like manner.

[0042] In the present invention, it is also possible to envisage stopping the temporal integration of the lateral spacing error by the calibration function if: - if the integral of the calibration function determined by the time integration of the lateral separation error or determined by the time integration of the calibration function corresponds to a predetermined threshold value, - if the measured and / or predicted curvature values ​​of the road surface on which the vehicle is or will be traveling exceed a predetermined threshold value; - if the curvature value of the target trajectory exceeds a predetermined threshold, - if the product of the square of the vehicle speed at that time and the curvature value of the road surface exceeds a predetermined threshold, and / or - if the product of the square of the vehicle speed at that time and the curvature value of the target trajectory exceeds a predetermined threshold.

[0043] The contribution of the calibration value to the controlled variable can be limited to a predetermined threshold value by at least temporarily halting the time integration of the lateral separation error in the calibration function or limiting the total integration to this threshold value. The threshold value in this case represents a maximum amplitude and can be determined, for example, based on the sum of the maximum possible influences of all offsets on the steering angle target and the steering angle actual. In this way, advantageously, measures to avoid integrator windup effects in the calibration function are not required.

[0044] As an alternative to limiting the integrator value within a threshold, the calibration function can also include a low-pass filter, such as a PT-1 filter. This method also avoids the windup effect. Although the sensor offset cannot be completely compensated for, the centering accuracy achieved by the calibration function can be sufficient.

[0045] Additionally or alternatively, the time integration of the lateral spacing error by the calibration function can be stopped if the measured and / or predicted curvature value of the road surface on which the vehicle is traveling exceeds a predetermined threshold and / or if the curvature value of the target trajectory exceeds a predetermined threshold.

[0046] When a vehicle is cornering, the lateral spacing error resulting from possible errors in predictive control depends not only on the curvature offset error but also on the inaccuracy of the stored vehicle parameters, particularly the uncertainty of the vehicle's specific steering angle gradient. If there is an error in the vehicle's curvature predictive control, this can have the same effect as a disturbance to the vehicle's lateral feedback control, thereby partially causing additional lateral spacing errors. Such additional lateral spacing errors can cause the incorrect intervention of a calibration function, which can be advantageously avoided or reduced by temporarily suspending the calibration function or at least the time integration of the lateral spacing error by the calibration function.

[0047] The effect of errors in the curvature prediction control on the vehicle's travel or lateral position may depend, inter alia, on the vehicle speed. To take this speed dependency into account, it is possible to envisage halting the time integration of the lateral spacing error by the calibration function when the product of the square of the current vehicle speed and the curvature of the road surface exceeds a predetermined threshold and / or when the product of the square of the current vehicle speed and the curvature of the target trajectory exceeds a predetermined threshold.

[0048] In a preferred embodiment, if a steering intervention by the vehicle driver generates a steering moment that exceeds a predetermined threshold and / or generates a steering moment that increases the current lateral spacing error, the time integration of the lateral spacing error according to the present invention is temporarily stopped and / or the application of the calibration value to the controlled variable is suspended. In this manner, lateral spacing errors caused by intentional driving operations that are not due to sensor offsets or similar factors, i.e., factors that must be corrected, are not reflected in the calibration value calculated by the calibration function, thereby advantageously preventing the lateral spacing errors from being reflected in the calibration value calculated by the calibration function. In other words, lateral deviations from the target trajectory intentionally caused by the driver, not due to lane guidance, can be advantageously prevented from affecting the calibration value.

[0049] In the present invention, if the driver of the vehicle performs a steering intervention that generates a steering moment that reduces the current lateral spacing error, the time integration of the lateral spacing error is carried out with an increased weighting, particularly preferably with an increased integration constant, at least for a predetermined time period.

[0050] If the driver intervenes by himself to generate a steering moment acting in the same direction as the calibration value generated by the calibration function, this can be interpreted as confirmation of the calibration function's adaptation direction. In this case, using an increased weighting or an increased integration coefficient can advantageously improve the dynamic characteristics of the improved calibration by the calibration function. From the driver's perspective, even if he cancels the corrective steering moment, it will appear that he has shifted the vehicle's course laterally more toward the center of the lane and will then maintain that course.

[0051] In one preferred embodiment, multiple integral functions can be used, each of which is assigned a different speed interval, and each integral function integrates the lateral spacing error over time at the vehicle speed within the speed interval assigned to it, and the calibration value of the integral function corresponding to the speed interval that includes the vehicle's current speed, or a total calibration value calculated based on the calibration value, is applied to the control variable.

[0052] As mentioned above, the influence of the offset on the lateral spacing error depends in part on the vehicle speed. This is valid, for example, for offset errors contained in the output of sensor disturbance compensation or offset errors that are components of curvature predictive control. To minimize the lateral spacing error, the calibration value applied to the controlled variable can be determined as a function of speed. In particular, due to the high integration time constant and the resulting low dynamics, constantly adapting the calibration value of this unique calibration function to changing vehicle speeds takes a relatively long time, which can potentially reduce the accuracy of the lane guidance.

[0053] To address this situation, it is advantageous to employ multiple calibration functions rather than using only one calibration function for the entire speed range of the vehicle. Each calibration function then determines its own calibration value, which can be stored independently of the other calibration values. By storing the calibration values ​​individually, for example in a non-volatile storage means, these values ​​can be expanded over time by integration during ongoing operation, or the entire state of the multiple calibration functions can be initialized individually when the vehicle is started.

[0054] In this case, the entire range of possible vehicle speeds can be preferably divided into several subsections, each covering a limited speed range and each assigned to its own calibration function. The controlled variable is then applied with the calibration value of the calibration function assigned to the speed section within which the current vehicle speed falls. Thus, only one calibration function is active at any one time, while the time integration of the lateral spacing error by the remaining calibration functions, in particular the other calibration functions, is stopped or a zero input signal is applied to their inputs. Meanwhile, the current lateral spacing error can be input to the currently active calibration function. The weighting of the lateral spacing error can be performed using a constant weighting factor or integration coefficient or a weighting factor or integration coefficient that depends on the vehicle speed. Subsequently, the calibration value of the integration function whose speed section includes the current vehicle speed, or the total calibration value determined depending on the calibration value, is applied to the controlled variable. The advantage of using multiple calibration functions or a bank of calibration functions over using only one calibration function is most apparent during acceleration or braking and when these frequently change speed ranges.

[0055] According to the invention, the total calibration value can be determined from the calibration values ​​generated by two or more integral functions in a time-continuous manner even when switching between two or more speed ranges, thereby avoiding abrupt changes in the calibration values ​​applied to the controlled variable and the associated abrupt interventions when switching between the calibration functions due to changes in vehicle speed.

[0056] In a preferred embodiment, the calibration functions can each determine the integral value of the calibration function attributed to them by integrating the lateral distance error over time, where the integral value of one or more integral functions whose current speed is outside each speed interval can be continuously adjusted, particularly preferably using a constant rate or a rate dependent on the vehicle speed, relative to the integral value of the calibration function whose current speed is within that speed interval.

[0057] In cases where the automatic lane guidance function is primarily used within a narrow speed range, only a few or only one calibration function is loaded for the lateral spacing error, thereby avoiding the current offset situation being reflected solely. To avoid this, inactive calibration functions not used at the current speed can continue to integrate the current lateral spacing error over time. In this case, the integral value, i.e., the integrator state of each calibration function assigned to a range adjacent to the current speed range, is limited to a maximum value corresponding to the integral value of the calibration function assigned to the current speed range. Furthermore, since the advantage of dividing the integrator state into multiple integrator states would be lost if the integrator coefficients were not changed, the weighting factors or integrator coefficients of calibration functions assigned to speeds other than the current speed range can be advantageously reduced compared to when they are active. The degree of reduction of the adjacent integrator coefficients depends on the expected maximum sensor drift and is, for example, in the range of 10% to 20%. This makes it possible to avoid a temporary increase in lateral spacing error and a decrease in driving comfort when entering an adjacent speed range, where the state of an inactive calibration function gradually deviates from the current situation due to typical drifts in curvature, yaw rate, and lateral acceleration offsets, until adaptation to the current offset situation is complete.

[0058] The continuous adjustment of the integrals of one or more integral functions where the current speed is outside the respective speed interval, particularly preferably with a constant rate or a rate dependent on the vehicle speed for the integrals of the calibration functions where the current speed of the vehicle is outside the speed interval, can also be considered as implementing a forgetting factor. In this way, the state of the inactive calibration functions can be brought closer to the state of the currently active calibration functions, for example with a constant gradient or a gradient set individually for each speed range.

[0059] The control device according to the present invention for generating a control variable for at least one lateral guidance actuator of a vehicle is a control device configured to perform the method according to one of the claims. The control device can also be configured to perform lateral feedback control, to determine a target trajectory, and / or to perform curvature and predictive control. Alternatively, the control device can perform only some of these functions. The control device can be connected to at least one sensor of the vehicle, in particular to at least one periphery sensor that at least partially captures at least one vehicle periphery and / or to at least one speed sensor that captures the vehicle speed, and / or to a navigation system of the vehicle.

[0060] A vehicle according to the invention can be envisaged that includes at least one lateral guidance actuator and a control device according to the invention, which is configured to control the at least one lateral guidance actuator using the control variable, which can be, for example, a steering actuator, particularly preferably an electric servo drive for the front or rear axle steering.

[0061] It is envisaged that the computer program according to the invention comprises instructions for causing a control means to carry out the method according to the invention, which control means can particularly preferably be a control device connectable to at least one lateral guidance actuator of the vehicle.

[0062] All advantages and embodiments described above with respect to the method according to the invention also apply to the control device according to the invention, to the vehicle according to the invention and to the computer program according to the invention, and vice versa.

[0063] Further advantages and details of the invention will become apparent from the following description of the embodiments and the drawings, in which: [Brief explanation of the drawings]

[0064] [Figure 1] FIG. 1 shows one embodiment of a vehicle according to the present invention. [Figure 2] FIG. 2 is a block diagram of an embodiment of a method according to the invention for implementing lane guidance in a vehicle; [Figure 3] FIG. 3 is a block diagram of the calibration functions of an embodiment of the method according to the invention. DETAILED DESCRIPTION OF THE INVENTION

[0065] 1 shows one embodiment of a vehicle 1. The vehicle 1 can be, for example, a motor vehicle such as a car or a truck. Subsequently, the vehicle 1 can also be, for example, a train consisting of a towing vehicle and one or more towed vehicles. Alternatively, the method can be employed with other types of vehicles 1.

[0066] The vehicle 1 comprises a control means 2 configured to implement the method for lane guidance of the vehicle 1. Subsequently, the vehicle 1 also comprises at least one lateral guidance actuator 3 with which the lateral position of the vehicle 1 can be set.

[0067] The lateral guidance actuator 3 can be embodied, for example, as a steering actuator with which the front axle angle of the vehicle 1 can be automatically set. The steering actuator can be embodied, for example, as an electric servomotor. Additionally or alternatively, the or other lateral control actuator 3 can be a steering actuator for automatically setting the rear axle steering angle.

[0068] The control means 2 is configured such that a control variable is determined via a lateral feedback control based on a lateral spacing error describing a lateral offset 4 of the vehicle 1 relative to a predetermined target trajectory 5 (shown by a dotted line), and the lateral position of the vehicle 1 can be adjusted to the target trajectory 5 depending on the control variable. To this end, the control means 2 controls the lateral guidance actuators 3 using the control variable. The lateral offset 4 describes, for example, a deviation in the lateral direction of the vehicle 1 between the target trajectory 5 relative to the center of the vehicle and the actual running direction 6 of the center of the vehicle.

[0069] The deviation from the target trajectory can then be determined within a defined distance (prediction distance) from the front end of the vehicle 1, as shown in Figure 1, or from another reference point. Further possible reference points can be, for example, the height of the rear or front axle of the vehicle 1. The reference point is preferably located on the longitudinal axis of the vehicle, but other fixed reference points during the observation period of the vehicle 1 are also conceivable.

[0070] 2 shows a block diagram of an embodiment of a method for implementing automated lane guidance of a vehicle 1, which is executed by a control means 2. The automated lane guidance comprises a lateral feedback control 7 for feedback controlling the lateral position of the vehicle 1 in order to maintain a target trajectory 4. The lateral feedback control 7 generates a control variable delta_set for the lateral guidance actuator 3, which can be controlled, for example, directly or via an intermediate total steering angle feedback control means (not shown).

[0071] To generate the control variable delta_set, the lateral feedback control 7 includes a path-tracking feedback control means 8, a curvature predictive control 9, one or more disturbance compensation means 10 for, e.g., crosswind, road tilt, and / or other effects, and a steering angle offset calibration 11. Using the path-tracking feedback control means 8, a control variable component delta_regler is determined.

[0072] The curvature predictive control 9 determines a predicted control variable delta_vorst depending on the measured and / or predicted curvature value of the lane on which the vehicle 1 is traveling, and the control variable delta_soll contains the predicted control variable delta_vorst as a control variable component. Similarly, one or more disturbance compensation means 10 generate one or more control variable components delta_komp, and the steering angle offset calibration 11 generates a control variable component delta_offset.

[0073] Subsequently, at least one calibration function 12 is provided, which determines a calibration value delta_kalib, which is applied to the control variable delta_soll generated by the lateral feedback control 7. The yaw rate g and the lateral acceleration a_lateral of the vehicle 1 are transmitted as input variables from the signal processing 13 to the at least one disturbance compensation means 10. For this purpose, the signal processing 13 can evaluate measured values ​​representing the yaw rate and / or the lateral acceleration detected by a number of sensors on the vehicle 1.

[0074] The calibration function 12 and the trajectory tracking feedback control means 8 are supplied with a lateral spacing error delta_y from a position recognition unit and trajectory planning unit 14. The lateral spacing error delta_y describes the lateral offset 4 between the target trajectory 5 and the actual direction of movement 6 or actual position of the vehicle 1 shown in FIG.

[0075] The position recognition unit and trajectory planning unit 14 subsequently provides the curvature prediction control 9 with at least one measured and / or predicted curvature information kappa describing the curvature of the target trajectory 5 within a section of the target trajectory 5 ahead in the direction of travel of the vehicle 1. The position recognition unit and trajectory planning unit 14 determines the target trajectory 5, and likewise the curvature information kappa, depending on surrounding data provided by a surrounding recognition unit 15 of the vehicle 1. The surrounding recognition unit 15 particularly preferably generates the surrounding data depending on surrounding measurement data generated by a surrounding sensor (not shown) capturing at least one partial area around the vehicle 1.

[0076] All or part of the functions of blocks 8 to 15 can be implemented in the control means 7. The output values ​​of blocks 8 to 11 all constitute a controlled variable delta_soll, which is subsequently corrected by at least one calibration value delta_kalib of a calibration function 12.

[0077] The individual components of the control variable delta_soll and the at least one calibration value delta_kalib have different effects on the lane guidance of the vehicle 1, as will be described below.

[0078] The target steering angle component delta_vorst of the curvature predictive control 9 is, for example, (1) delta_vorst=kappa*(l+EG*v^2), In the formula, the parameter l corresponds to the axle distance, and EG corresponds to the inherent steering angle gradient of the vehicle 1. The inherent steering angle gradient is expressed as follows: (2) EG=m*(ch*lh-cv*lv) / (ch*cv*(lh+lv)) As shown above, it depends on the distance Iv from the center of gravity of the vehicle 1 to the front axle, the distance Ih from the center of gravity of the vehicle 1 to the rear axle, the vehicle weight m of the vehicle 1, and the cornering stiffness cv of the front wheels and the cornering stiffness ch of the rear wheels.

[0079] When the curvature information kappa has an offset kappa_offs due to an error compared with the actual curvature, the predictively controlled target steering angle is also (3)delta_vorst_kappaoff=kappa_offs*(l+EG*v^2) The result of the predictive control delta_vorst, including the error due to delta_vorst_kappaoff, acts, so to speak, like a disturbance to the lateral feedback control 7, such that when the lane keeping assistant is activated, the vehicle 1 does not follow the target trajectory 5 as expected, but instead, depending on the specifications of the trajectory following feedback control means 8—i.e., depending on the type of feedback control used, with or without steady-state accuracy—the vehicle 1 exhibits a continuous lateral offset 4 relative to the planned target trajectory 5, or at least causes a temporary deviation, until the vehicle 1 follows the planned target trajectory 5 after the adjustment process has decayed.

[0080] The degree of image distortion of the camera used as the periphery sensor supplying the periphery measurement data to the position recognition unit and trajectory planning unit 14 is usually constantly changing during the operation time of the driver assistance system due to the camera calibration routines that are typically running continuously. This means that depending on the state of the calibration, the value of the offset of the curvature predictive control 9, and therefore the degree of the negative impact on the lateral guidance of the vehicle, will change over time, even if the change in the transfer characteristic of the camera and the associated curvature offset is at least not rapid and band-limited.

[0081] The steering angle can also be used as an auxiliary variable or feedback control value in a lane guidance and driver assistance system. The steering angle can be influenced or corrected by a steering angle calibration routine 11. The determined steering angle offset delta_offset typically has an uncompensated component due to the finite accuracy of the steering angle calibration routine 11.

[0082] The residual offset error delta_offset_error in the steering angle signal can have several possible causes. Typically, the steering angle calibration routine 11 relies on the signals of the vehicle 1's yaw rate sensor and / or the vehicle's wheel rotation speed sensors. Consequently, errors from these sensors, such as offset errors and linearity errors, can be inherited by the calculated steering angle offset. The effect of the steering angle offset, similar to the curvature offset described above, is that the vehicle 1 does not follow the target trajectory 5 as expected, but instead exhibits a persistent lateral offset 4 relative to the planned target trajectory 5, which, depending on the specifications of the trajectory tracking feedback control means 8, can adversely affect the comfort and acceptability of the driver assistance system.

[0083] The trajectory tracking feedback control means 8, for example, (4)Gr(s)=Kp(v)+s*Kd(v) / (1+s*T), and when using a PD feedback control means with speed-dependent feedback control factors Kp(v) and Kd(v), When the lateral spacing error delta_y describing the lateral offset 4 is an input value, the output value of the trajectory tracking feedback control means 8, delta_regler, is expressed as follows: (5)delta_regler(s)=Gr(s)*delta_y(s) is obtained by

[0084] As a result, the path tracking feedback control means 8 is in a steady state of equilibrium during straight-ahead driving, i.e., when the physical steering angle is zero degrees and lateral disturbances acting on the vehicle 1 are ignored, it outputs an uncorrected steering angle offset delta_offset_error. The lateral offset 4 required for this is (6)Delta_y(0)*Gr(0)=Delta_offset_error(0) and in the time domain, (7)delta_y=delta_offset_error / Kp(v) Since the feedback control factor Kp cannot be chosen arbitrarily large during design, a lateral offset 4 of the vehicle 1 relative to the target trajectory 5 is always formed when the uncorrected steering angle offset delta_offset_error is not zero. If the vehicle 1 is rear-axle steered, the uncorrected offset of the rear-axle steering angle then has a comparable effect on the lateral offset 4 of the vehicle 1 as the uncorrected offset of the front-axle steering.

[0085] The sensor signals for the yaw rate, the longitudinal steering angle and the lateral acceleration are also used to estimate disturbance forces and load moments acting on the vehicle 1, which are the result of, for example, the inclination of the road surface or a crosswind acting on the vehicle 1. The determined forces and moments contribute proportionately to the steering angle setpoint delta_soll by implementing the disturbance compensation delta_komp.

[0086] Subsequently, the sensors can be used to estimate the slip angle of the vehicle 1, which can also be used as an auxiliary variable for load estimation. It is observed that offset errors in the yaw rate g, steering angle, or lateral acceleration a_lateral frequently have an effect on the estimated values, proportional to the respective offsets, in the steady-state case, and thus also on the disturbance compensation. As a result, the yaw rate g, steering angle, and lateral acceleration a_lateral are also passed on to the disturbance compensation components of the steering angle setpoint delta_soll. As a result, the lane-guided vehicle 1 again experiences undesirable temporary or persistent lateral deviations from the planned target trajectory 5. Disturbance compensation means 10 for the yaw rate g and lateral acceleration a_lateral, like the steering angle correction 11, usually only have a finite accuracy and therefore may lag behind the actual offsets. Therefore, even with these compensation means, the aforementioned adverse effects on vehicle comfort can be observed when using lane guidance and driver assistance systems.

[0087] These effects are addressed and at least partially compensated for by at least one calibration function 12, thereby improving the ride comfort of the vehicle 1. In this regard, the calibration function 12 generates a calibration value delta_kalib that is superimposed on the lateral feedback control 7 and applied to at least one control variable delta_soll. The calibration value delta_kalib is determined by the at least one calibration function 12 at least partially as a function of the time integral delta_y of the lateral separation error.

[0088] Depending on the sign and amplitude of the lateral spacing error delta_kalib, a calibration function 12 determines a calibration value delta_kalib for the setpoint steering angle delta_soll, which is preferably adapted so that the lateral spacing error delta_y is always reduced. This calibration value delta_kalib is then added together with the control variable delta_soll to the control variable components of the trajectory tracking feedback control means 8, the curvature predictive control 9, one or more disturbance compensation means 10, and the steering angle offset calibration 11, and is output as a calibrated control variable delta_soll_kalib, for example, to the setpoint angle interface of the lateral guidance actuator 3. Alternatively, the calibrated control variable can also be output as an output to the moment interface of the lateral guidance actuator 3.

[0089] The calibration function 12 may be implemented, for example, as an integrating means or as an I element with a relatively small integral coefficient k, which integrates the lateral spacing error with small dynamic characteristics. Alternatively, it may be realized as the PT1 element of the calibration function or as another type of different function which exhibits at least partially integral behavior.

[0090] Therefore, the determination of the integral value delta_kalib by the calibration function 12 is carried out as an integrating means using the integral coefficient ki, for example, (8)Delta_kalib=Delta_y*ki / s It can be expressed as:

[0091] The time constant of the calibration function 12 can be set to be at least twice, at least five times, at least ten times, at least one hundred times or at least one thousand times greater than the dominant time constant of the lateral feedback control 7 or of the trajectory tracking feedback control means 9. The integral coefficient k is therefore, for example, of the order of 0.001 Grad / (m*s), and for the lateral spacing error delta_y considered as an illustrative example, the gradient of the calibration value 12 is only 0.0002 Grad / s from 0.2 m.

[0092] The adaptive dynamics of the calibration value 12 are preferably set to be comparable in magnitude to the dynamics of the change in the offset error. An indicator for this is, for example, the maximum expected offset drift due to the temperature of the sensors used in the lateral feedback control 7, converted, for example, into the corresponding steering angle plane. If the dynamics are selected too large, typical overshoots occur in the signal for delta_kalib, since the calibration partially competes with the convergence characteristics of the compensation function 10 for road cross gradients and crosswinds. However, this can limit the comfort improvement achieved by the calibration function 12.

[0093] The calibration function 12 calculates its integral value by integrating the lateral spacing error delta_y over time. This integral value represents the result of the time integration of the lateral spacing error delta_y at a specific point in time. This integral value is retained even when lane guidance is deactivated and is used as the initial value for a new time integration of the lateral spacing error delta_y when lane guidance is reactivated. The state of the calibration function 12, implemented for example as an integrator, is not reset during the current ignition cycle but is maintained even if the driver assistance is repeatedly deactivated and activated, e.g., after the end of the journey and / or due to driver intervention that overrides the automatic lane guidance.

[0094] Using a constant integral coefficient k i in the calibration function 12 may be a compromise for the entire speed range that the vehicle 1 can reach, since the effect of the offset on the lateral spacing error delta_y may depend on the driving speed, which may be due to, for example, the driving speed dependent controller parameters typically used in the tracking feedback control means 8. This will also change the disturbance compensation characteristics of the tracking feedback control means 8, and consequently the effect of the sensor offset on the lateral spacing error delta_y.

[0095] To address this situation and expand the potential of the calibration function, it is conceivable to implement a time integration of the lateral distance error delta_y using a weighting factor, particularly preferably an integration coefficient ki that depends on the vehicle speed. The integration coefficient ki can thus be implemented as ki(v), which depends on the current speed of the vehicle 1. This advantageously achieves individually adjustable calibration dynamics of the calibration function 12 that are optimized for the respective driving speed range.

[0096] Additionally or alternatively, it is conceivable that the weighting factor or integration coefficient k i increases with increasing absolute value of the lateral spacing error delta_y. The calibration speed of the calibration function 12 can be effectively improved by additional non-linear or progressive weighting of the absolute value of the lateral spacing error delta_y.

[0097] For example, the integral coefficient ki and therefore the adaptation rate can be switched between two or more values, with a larger value being selected for the integral coefficient ki when the absolute value of the lateral spacing error delta_y is larger. For example, if the absolute value of the lateral spacing error delta_y exceeds 0.2 m, the integral constant ki can be doubled to allow for a faster initial adaptation to the offset situation that exists when the driver assistance function is activated.

[0098] By defining the adaptation rate of the calibration function 12 as a function of the absolute value of the lateral separation error delta_y, it is possible to achieve a fast adaptation to new offset situations, even if the sensor compensator 10 and / or other sensor offset calibration functions in the vehicle 1 finally eliminate the detected sensor offset drift by step-wise correction of each offset only after an isolation time has elapsed. This is particularly advantageous when the isolation time of the sensor compensator 10 and / or other sensor offset calibration functions is relatively long, since during such isolation time the calibration function 12 will perform successive compensations which will then have to be integrated back again after a signal update of the sensor compensator or sensor offset calibration function.

[0099] Since a step-like correction of the sensor offset directly leads to a sudden lateral movement of the vehicle 1 depending on the height of the step, it is possible to assume, as an advantageous extension, that all sensor offset calibration functions in the vehicle 1 do not change their sensor offset in a step-like manner but only within a limited bandwidth. If the offset compensation for the on-board sensors is performed only with a defined maximum gradient, for example, and the centering calibration output delta_kalib can always follow this with only a small following error of about 0.03 degrees, the non-linear or gradual weighting of the absolute value of the lateral spacing error delta_y in the calibration function 12 can be omitted as a measure to improve dynamics.

[0100] Subsequently, the contribution of delta_kalib to the total control variable delta_soll_kalib is advantageously limited in its maximum amplitude by limiting the state of the integrating means in the calibration function 12. For this reason, it is provided that the integration over time of the lateral spacing error delta_y by the calibration function 12 is stopped when the integration over time of the lateral spacing error delta_y by the calibration function 12 reaches a predetermined threshold value.

[0101] The threshold or maximum amplitude of the integrator or calibration function 12 is then based, for example, on the sum of the maximum influences of all offsets on the steering angle target value delta_soll and the actual steering angle. Here, if delta_komp, delta_vorst, and delta_offset each have an offset error of, for example, 0.05 degrees, delta_kalib should be limited to a range of + / - 0.15 degrees. These angle expressions are based on the steering angle of the vehicle, for example, the steering angle of the vehicle's front wheels. This method advantageously eliminates the need for additional measures to avoid integrator windup effects in the calibration function 12.

[0102] As an alternative to limiting the integrator state, a low-pass filter, such as a PT-1 filter, can be used. This method also avoids the windup effect. Although offset situations cannot be completely eliminated, the centering accuracy can be sufficient.

[0103] When cornering, the lateral spacing error resulting from the error of the predictive control depends not only on the curvature offset error but also on the inaccuracy of the stored vehicle parameters, especially the uncertainty of the intrinsic steering angle gradient EG. The error component of the predictive control due to the error of the intrinsic steering angle gradient EG can be calculated by applying total differentiation to equation (1) and evaluating it for EG or its error EG_err. (8)delta_vorst_EG_err=(d delta_vorst / d EG)*EG_err =(d(kappa*(l+EG*v^2)) / d EG)*EG_err =kappa*v^2*EG_err is obtained as:

[0104] The contribution of delta_vorst_EG_err acts as a disturbance to the lateral feedback control of the vehicle 1 and proportionally causes the lateral spacing error delta_y, which results in an incorrect steering of delta_kalib. That is, the signal delta_kalib does not only react to the sensor offset. As long as EG_err is limited, the effect of EG_err on the vehicle level is small due to the large integral time constant of the centering calibration. In this case, the integral time constant is defined as the reciprocal of the integral coefficient ki.

[0105] This is also true because incorrect steering of delta_kalib in a curve is corrected again in the straight part of the lane.To counteract significant variations in EG_err, an adaptive algorithm can continuously adjust the EG value in the vehicle to the current conditions, such as the vehicle mass m and tire stiffness, which reduces EG_err and therefore the contribution of delta_vorst_EG_err.

[0106] If the fluctuation of EG_err continues to be significant and the product of kappa * v^2 takes a large value, the calibration function 12 can be interrupted. That is, if the product of the square of the vehicle speed v at that time and the curvature value kappa of the target trajectory 5 and / or the curvature value of the lane exceeds a predetermined threshold, it can be assumed that the temporal integration of the lateral distance error delta_y by the calibration function 12 will be interrupted.

[0107] For this purpose, the curvature value kappa is low-pass filtered, for example by means of the PT-1 algorithm, and then its absolute value is compared with a threshold value that depends on the driving speed v. If the filtered curvature exceeds this threshold value, the calibration function 12 is temporarily stopped.

[0108] It is also conceivable that the time integration of the lateral spacing error delta_y by the calibration function 12 is subsequently stopped if the measured and / or predicted curvature value kappa of the road surface on which the vehicle 1 is traveling exceeds a predetermined threshold value and / or if the curvature value of the target trajectory 5 exceeds a predetermined threshold value.

[0109] If the curvature itself temporarily has a large offset error, this may reduce the calibration work of the high-speed calibration function 12 until the camera calibration compensates for this error. However, the adaptation region then shifts only by the amount of the curvature offset, so the calibration work does not stop completely. Furthermore, when the driving speed v decreases, and kappa * v^2 decreases accordingly, the calibration is again performed strongly.

[0110] Calibration disturbances can also be caused by driver interventions, which can have a direct effect on the lateral spacing error delta_y, for example by steering the steering wheel of the vehicle 1. To counter this, the calibration function 12, i.e. the integration of the lateral spacing error delta_y, can be suspended in the event of driver interventions at the steering wheel. This means that the integration of the lateral spacing error delta_y over time is suspended and / or the application of the calibration value delta_kalib to the control variable delta_soll is suspended in the event of a driver intervention at the steering wheel that generates a steering moment M that exceeds a predetermined threshold and / or generates a steering moment M that increases the current lateral spacing error delta_y.

[0111] However, improved availability of the calibration function 12, and therefore faster convergence, can be achieved by stopping the calibration function 12 only when the driver steers in a direction that increases the lateral spacing error delta_y. This can be done by comparing the signs of the driver steering moment M and the lateral spacing error delta_y. For example, if the counting arrow of the driver steering moment M is defined so that steering to the left in the direction of travel produces a positive moment, and a positive lateral spacing error means that the vehicle is located to the right of the planned target trajectory 5 in the direction of travel, then the stopping condition for the calibration function must be that the signs of the driver steering moment M and the lateral spacing error are different.

[0112] The integration coefficient ki is first estimated, for example, using the magnitude of the measured driver steering moment M, and this is used as the basic specification for implementing the lane keeping assistant when the driver's steering intervention is less than a predetermined intensity. Additionally or alternatively, if the driver of vehicle 1 performs a steering intervention that generates a steering moment M that reduces the current lateral spacing error delta_y, the time integration of the lateral spacing error delta_y is performed with an increased weighting, particularly preferably with an increased integration constant, at least for a predetermined time period. This makes it possible to intentionally increase the integration coefficient above the basic specification value when a driver intervention is detected and the driver is steering in a direction that reduces the lateral spacing error delta_y. In this case, since the driver's steering can be considered as acceptance of the calibration process by calibration function 12, increasing the integration coefficient ki and thereby increasing the calibration dynamics are justified.

[0113] As mentioned above, the effect of the offset amount on the lateral spacing error delta_y depends in part on the vehicle speed v. This is valid, for example, for the offset error contained in the output delta_komp of the disturbance compensation means or the offset error that is a component of the predictive curvature control delta_vorst. Therefore, when the vehicle speed changes, delta_kalib must be constantly changed to keep the lateral spacing error small, even if the offset values ​​of the yaw rate g, lateral acceleration a_lateral, and predictive curvature kappa, which contribute to the lateral spacing error, are assumed to be constant. Even if the integration time constant of the calibration function 12 is high and therefore the dynamics of the calibration function 12 are low, it takes a relatively long time to adapt to a changed vehicle speed v, during which time the lane-following accuracy decreases.

[0114] To address the above situation, rather than using a single integrator means for storing delta_kalib for the entire speed range of the vehicle, it is advantageous to employ an arrangement of multiple integrators or storage means, as illustrated in FIG. 3.

[0115] The use of multiple integration functions 12_1 to 12_N is illustrated in Figure 3. Furthermore, an interpolation unit 16, an initialization unit 17, an integration coefficient determination unit 18, and a sequence control unit 19 are depicted as components of the control means 2.

[0116] Each of the integrating functions 12_1 to 12_N is assigned a different speed interval. Each of the integrating functions 12_1 to 12_N is configured to integrate the lateral spacing error delta_y over time when the vehicle 1 is traveling within the assigned speed interval. The entire vehicle speed range is divided into N>1 subintervals, each of which covers a limited speed range and is assigned a dedicated calibration function 12. Therefore, only one calibration function 12 is active at any given time; the remaining calibration functions 12 are inactive or have an input signal of zero applied to their input terminals. The weighted lateral spacing error delta_y is input to one of the active calibration functions 12_1 to 12_N, and each integrating means generates a corresponding delta_kalib_v(k), k=[1...N]. The corresponding calibration value delta_kalib_v(1) to delta_kalib_v(N) of one of the calibration functions 12_1 to 12_N, each speed interval covering the current speed of the vehicle 1, applies the resulting delta_kalib directly to the control variable delta_soll. The weighting of the lateral spacing error delta_y can then be carried out using a constant integral coefficient ki or a coefficient ki(v) that depends on the vehicle speed.

[0117] The integral coefficients k i to k i(v) of the individual calibration functions 12_1 to 12_N can be determined by an integral coefficient determination unit 18. For this purpose, the integral coefficient determination unit 18 can receive as input values, for example, the vehicle speed v, the steering moment M of the driver of the vehicle 1, the current lateral spacing error delta_y, as well as the measured or predicted curvature kappa.

[0118] The centers of the N speed intervals essentially form nodes of the output value delta_kalib_v(k). To obtain the resulting delta_kalib from the individual contributions delta_kalib_v(k), an overall calibration value delta_kalib_ges determined by an interpolation unit 16 from the calibration values ​​delta_kalib_v(1) to delta_kalib_v(N) can additionally or alternatively be applied to the controlled variable delta_soll. The overall calibration value delta_kalib_ges is then determined from the calibration values ​​delta_kalib_v(1) to delta_kalib_v(N) generated by two or more integration functions 12, preferably such that the overall calibration value delta_kalib_ges is continuous even when switching between two or more speed intervals. This can be achieved, for example, by linear interpolation by the interpolation unit 16, which receives the vehicle speed v as input.

[0119] In this case, delta_kalib_ges is a linear combination of delta_kalib_v(k) and delta_kalib_v(k+1), depending for example on the current speed and on the distance to the center of the adjacent speed range, i.e. (9)delta_kalib_ges=delta_kalib_v(k)*a+delta_kalib_v(k+1)*(1-a) However, in the formula: a=(vv(k)) / (v(k+1)-v(k)),v>v(k),v <v(k+1) This becomes:

[0120] Interpolation between the individual integrator states delta_kalib_v(k) is also advantageous because it prevents step-like discontinuous changes in delta_kalib_ges when transitioning from one speed range to another. Steps in the steering angle setpoint delta_soll_kalib would cause the vehicle 1 to move abruptly sideways and must be avoided from the perspective of ride comfort. The advantages of using a bank of calibration functions 12_1 to 12_N or a bank of integrators over using only one integrator are particularly evident during acceleration or braking and when these result in frequent changes in speed ranges.

[0121] If the driver assistance function is mainly active within one speed range, it may happen that only one of the calibration functions 12 is provided with the lateral spacing error delta_y and that only this calibration function 12 correctly reflects the offset situation at that time. Due to typical drifts in the offset values ​​for curvature kappa, yaw rate g and lateral acceleration a_lateral, the states of the remaining integrating means will gradually become out of sync with the current situation.

[0122] When entering an adjacent speed range, the lateral spacing error delta_y temporarily increases and the ride comfort deteriorates until adaptation to the current offset situation is completed. A forgetting factor can be advantageously implemented as a remedy. For this purpose, it is conceivable that each of the calibration functions 12_1 to 12_N determines the integral value of the calibration function 12_1 to 12_N attributed to it by integrating the lateral spacing error over time, whereby the integral value of one or more of the integration functions whose current speed is outside the respective speed interval can be continuously adjusted, particularly preferably by a constant or vehicle speed-dependent factor relative to the integral value of the calibration function 12_1 to 12_N whose current speed of the vehicle 1 is within that speed interval.

[0123] For example, all integrator states or integral coefficients of inactive calibration functions 12_1 to 12_N are asymptotically adjusted from the state or integral value of the currently active calibration function 12_1 to 12_N with a uniform or individually predetermined gradient for each speed range. This gradient can be set, for example, depending on the dynamic characteristics of the offset value. In this application, it is advantageous to estimate that 25% of the gradient set for delta_kalib_v(k) corresponds to a lateral spacing error of 0.2 m. If a leveling gradient is selected that is too steep, the advantage of using a bank of calibration functions 12_1 to 12_N instead of using only one calibration function 12 is lost.

[0124] An alternative to the gradient method described above is to continue performing the calibration function for the currently irrelevant speed ranges with an integrator coefficient reduced by, for example, 10% to 20%, with this approach preventing the integrator state from exceeding the integrator state for the currently operating speed range by a corresponding limit.

[0125] Depending on the dynamic characteristics of the offset influence, it may be advantageous to store the state of the individual calibration function 12 or the states of the calibration functions 12_1 to 12_N in a non-volatile storage means, for example a storage means of the control means 3, so that the already calibrated state of the calibration function(s) 12 to 12_1 to 12_N can be set upon restart of the vehicle 1. This is particularly advantageous if the majority of the offset errors have no or only small drift over time.

[0126] The saved states or integral values ​​of one or more calibration functions 12_1 to 12_N can be implemented by an initialization unit 17. The initialization unit 17 performs the initialization of the calibration functions 12_1 to 12_N as well as the state comparison of the current integral values ​​of the calibration functions 12_1 to 12_N.

[0127] The flow control unit 19 can monitor events such as activating and / or deactivating the automatic lane guide, the occurrence of a condition for stopping the temporal integration, and / or the adaptation of at least one calibration value delta_kalib to the control variable delta_soll, switching between speed zones, and / or other events, and control the operation of at least one calibration function 12 accordingly.

[0128] In its basic function, the calibration function 12 can be considered as a form of integral component of the trajectory tracking feedback control means 8, but preferably has essential characteristics or differences, such as the integral coefficient or other weighting factor used for the time integration of the lateral spacing error delta_y being very small compared to the lateral feedback control 7, contrary to typical specification conditions, and / or the calibration function 12 or the time integration of the lateral spacing error delta_y is not initialized to zero when the assistance system is restarted, but is started or continued from the last reached state. The advantage of this is that when the driver assistance function is activated, the vehicle 1 can immediately travel along the planned target trajectory 5 without having to undergo a substantial lateral convergence process. The above-mentioned extension of the at least one calibration function 12 is a further difference from the conventional integral component of the lateral guidance feedback control means.

[0129] The at least one calibration function 12 can be implemented as an independent function or alternatively or additionally to an existing integral component in the tracking feedback control means 8. In this case, the at least one calibration function 12 is adapted only during the start-up phase of the driver assistance system, until the integral component of the tracking feedback control means 8 reduces the lateral spacing error delta_y equal to zero, or whenever the integral component of the tracking feedback control means 8 has to be reset and reconfigured. This can occur, for example, in a level 2 autonomous driving system when the driver intervenes in the steering.

[0130] Overall, the calibration dynamics of the calibration function 12, i.e., the time required for the lateral displacement 4 of the vehicle 1 to adjust to the current offset situation, is longer than in a driver assistance system without an integral component in the tracking feedback control means 8. This issue can be addressed, at least in part, by increasing the integral coefficient or coefficients when using one or more calibration functions 12 with a tracking feedback control means 8 that does not have an integral component. Furthermore, when an integral component is used in the tracking feedback control means 8, it is possible to omit one or more disturbance compensations 10 for crosswind and / or road slope. This eliminates one of the major sources of offset error, resulting in a reduction in the number of offset effects that the calibration function 12 must take into account, which further has the advantage of reducing the demands on the calibration dynamics of the at least one calibration function 12.

Claims

1. 1. A method for implementing lane guidance for a vehicle (1), in which a control variable is determined via a lateral feedback control (7) based on a lateral spacing error describing a lateral deviation (4) of the vehicle (1) relative to a predetermined target trajectory (5), and the lateral position of the vehicle (1) is adjusted to the target trajectory (5) in response to the control variable, comprising: and applying at least one calibration value to the controlled variable, the calibration value being determined by at least one calibration function (12) at least partly depending on the time integral of the lateral spacing error. Characterized by A method for implementing lane guidance.

2. The time constant of the calibration function (12) is particularly preferably at least twice as large as the dominant time constant of the lateral feedback control (7). Characterized by The method of claim 1.

3. The integral value of the calibration function (12) is calculated by integrating the lateral distance error over time, and this integral value is maintained even when the lane guide is stopped, and is used as the initial value of the lateral distance error over time integration when the lane guide is restarted. Characterized by The method according to claim 1 or 2.

4. The lateral feedback control (7) includes a predictive control (9), wherein the predictive control (9) determines a predictive control amount depending on a measured and / or predicted curvature value of the road surface on which the vehicle (1) is traveling, and the predictive control amount includes the predictive control amount as one of its components. Characterized by 10. A method according to one of the preceding claims.

5. The predictive control is performed depending on a vehicle model describing the vehicle's inherent steering angle gradient, which gradient is adapted during operation of the vehicle depending on at least one vehicle parameter, in particular the vehicle weight and / or the tire stiffness. Characterized by The method of claim 4.

6. The integration of the lateral spacing error over time is carried out using a weighting factor, particularly preferably an integration coefficient that increases depending on the speed of the vehicle (1) and / or with an increase in the absolute value of the lateral spacing error. Characterized by 10. A method according to one of the preceding claims.

7. The time integration of the lateral spacing error by the calibration function (12) is stopped if Characterized by A method according to one of the preceding claims: - if the integral of the calibration function (12) determined by the time integration of the lateral distance error or determined corresponds to a predetermined threshold value, - if the measured and / or predicted curvature values ​​of the road surface on which the vehicle (1) is or will be traveling exceed a predetermined threshold value, - if the curvature value of the target trajectory (5) exceeds a predetermined threshold value, - if the product of the square of the current vehicle speed and the curvature value of the road surface exceeds a predetermined threshold value, and / or - if the product of the square of the current vehicle speed and the curvature value of the target trajectory (5) exceeds a predetermined threshold value.

8. When a steering intervention by the driver of the vehicle (1) generates a steering moment exceeding a predetermined threshold and / or generates a steering moment that increases the current lateral distance error, the integration of the lateral distance error over time is temporarily stopped and / or the application of a calibration value to the control variable is suspended. Characterized by 10. A method according to one of the preceding claims.

9. If the driver of the vehicle (1) makes a steering intervention that generates a steering moment that reduces the current lateral spacing error, the integration of the lateral spacing error over time is performed with increased weighting, particularly preferably with an increased integration constant, at least for a predetermined time period. Characterized by 10. A method according to one of the preceding claims.

10. A plurality of integral functions (12) are used, each integral function (12) being assigned a different speed section, each integral function (12) integrating the lateral distance error over time at the speed of the vehicle (1) within the speed section assigned to each integral function (12), and a calibration value of the integral function (12) corresponding to the speed section including the current speed of the vehicle (1), or a total calibration value calculated based on the calibration value, is applied to the control variable. Characterized by 10. A method according to one of the preceding claims.

11. The total calibration value is determined from the calibration values ​​generated by two or more integral functions (12) so as to be continuous in time even when switching between two or more speed intervals. Characterized by The method of claim 10.

12. The plurality of calibration functions (12) each determine an integral value of the calibration function (12) attributed to the calibration function (12) by integrating the lateral distance error over time, with the integral value of one or more integration functions whose current speed is outside the respective speed interval being continuously adjusted, particularly preferably by a constant rate or a rate dependent on the vehicle speed, relative to the integral value of the calibration function (12) whose current speed is within the respective speed interval. Characterized by 12. The method according to claim 10 or 11.

13. At least one calibration function (12) is implemented as an I element or a PT1 element, respectively. Characterized by 10. A method according to one of the preceding claims.

14. 1. A control device (2) for generating a control variable for at least one lateral guidance actuator (3) of a vehicle (1), characterized in that the control device (2) is configured to implement a method according to one of the preceding claims.

15. 15. A vehicle comprising a lateral guidance actuator (3) and a control device (2) according to claim 14, characterized in that the control device (2) is adapted to control at least one of the lateral guidance actuators (3) using a control variable.

16. A computer program comprising instructions for prompting a control means to carry out the method according to claims 1 to 14.

Images