Method and driver assistance system for parking assistance of a motor vehicle

Abstract

The invention relates to a method and a driver assistance system for parking assistance of a motor vehicle. According to the method for parking assistance of a motor vehicle (1) in a manual steering parking maneuver, sensor data representing the environment of the motor vehicle (1) are generated by means of an environment sensor system (3) of the motor vehicle (1); at least one reference trajectory (7) is determined by means of a computing unit (4) of the motor vehicle (1) from the sensor data, and actual values of parameters which are variable at a point in time during the parking maneuver are determined by means of a sensor system (5) of the motor vehicle (1). At least one predicted deviation of the parameters from theoretical values is determined by means of the computing unit (4) from the at least one reference trajectory (7) and from the actual values, and a haptic signal is automatically generated as a steering recommendation for the driver of the motor vehicle (1) from the at least one predicted deviation.

Description

Technical Field

[0001] This invention relates to a method for parking assistance of a motor vehicle in manual steering parking operations, wherein sensor data representing the environment of the motor vehicle is generated using the vehicle's environmental sensor system; at least one reference trajectory of the motor vehicle is determined based on the sensor data using the vehicle's computing unit; and the actual values ​​of parameters of the motor vehicle that can vary during the parking operation are determined at a specific point in time using the vehicle's sensor system by the computing unit. The invention also relates to a corresponding driver assistance system and a motor vehicle having such a driver assistance system. Background Technology

[0002] In assisted parking of motor vehicles, the vehicle's environmental sensor system can identify suitable parking spaces, and the driver assistance system can plan the optimal trajectory for maneuvering the vehicle into the parking space. While the driver manually drives the vehicle into the parking space, they receive instructions from the driver assistance system to follow the optimal trajectory as closely as possible. However, the driver needs time to perceive and understand the instructions from the driver assistance system and ultimately execute them. This results in the optimal trajectory not always being followed, or the optimal parking position or orientation not always achieved. The higher the vehicle speed and the longer the driver's reaction time during parking, the greater the tendency for deviation.

[0003] A method for outputting parking instructions is described in document DE 10 2004 047 484 A1. Here, a steering command to deflect the steering device is output, and if a preset steering angle is not achieved after a preset distance following the output of the steering command, a correction command is output. However, a drawback of this method is that the correction command may be given too late compared to achieving the optimal parking position without additional parking maneuvers. Furthermore, the driver must execute not only the steering command but also the correction command, thus the aforementioned problems are not resolved but merely shifted.

[0004] Document DE 10 2004 001 122 A1 describes a method for providing information for parallel parking of a vehicle. Here, entry into a parking space should be optimally achieved independently of the driver's entry speed and reaction time. To this end, optical or acoustic signals are output to the driver during the parking process to persuade the driver to stop the vehicle. This signal is output taking into account the driver's reaction time and vehicle speed to ensure timely stopping of the vehicle. Repeatedly stopping the vehicle during the parking process reduces comfort and prolongs the parking process beyond what is necessary. Summary of the Invention

[0005] In this context, the objective of the present invention is to provide an improved solution for parking assistance of a motor vehicle in manual steering parking operations, which enables the vehicle to reach a target position with greater reliability or accuracy without reducing driving comfort for the driver.

[0006] The task is addressed by the corresponding content of the independent claims. Advantageous extensions and preferred embodiments are the content of the dependent claims.

[0007] The improved approach is based on the idea of ​​generating tactile signals as steering suggestions for the driver, which are generated based on predicted or anticipated steering deviations rather than current ones.

[0008] According to this improved scheme, a method for parking assistance of a motor vehicle in manual steering parking operations is described. Sensor data representing the environment of the motor vehicle is generated using the vehicle's environmental sensor system. A computing unit of the motor vehicle determines at least one reference trajectory for the motor vehicle, particularly for parking operations, based on the sensor data. The computing unit determines, using the vehicle's sensor system, the actual values ​​of parameters of the motor vehicle that can vary during the parking operation at a given time point. The computing unit determines, based on the at least one reference trajectory and the actual values, at least one predicted deviation between the parameters and theoretical values ​​for those parameters. Based on the at least one predicted deviation, particularly using the vehicle's actuator unit, a tactile signal is automatically generated as a steering suggestion for the driver of the motor vehicle.

[0009] Manual steering parking can be understood in particular as not involving fully automatic lateral control during parking. Specifically, the driver must always manually or partially manually intervene in the steering to bring the vehicle to a stop according to the parking maneuver. Longitudinal control of the vehicle can be performed entirely manually, partially automatically, or fully automatically.

[0010] Currently, the reference trajectory does not necessarily have to be a theoretical trajectory for motor vehicles, i.e., a trajectory that reflects the optimal driving path for parking. The at least one reference trajectory may include such a theoretical trajectory. However, alternatively or additionally, the at least one reference trajectory may also include other trajectories suitable for correcting actual values. For example, this could be a theoretical trajectory modified in an appropriate manner.

[0011] Environmental sensor systems may include, in particular, ultrasonic sensor systems, one or more cameras, one or more radar systems, and / or one or more lidar systems.

[0012] The parameters that can vary during parking maneuvers are those that directly or indirectly relate to the movement of the vehicle during the parking maneuver. These parameters may include, for example, the vehicle's position, orientation, steering angle, or the curvature of the trajectory traversed by the vehicle. Multiple different measurements can also be combined or processed to determine the parameters.

[0013] Prediction deviation of parameters can be, in particular, the deviation between the estimated or predicted parameter and the theoretical value at a time point after the actual value is determined. For this purpose, parameters can be predicted, for example, based on a predicted trajectory for a motor vehicle, and the predicted value can be compared with the theoretical value. However, other possibilities for prediction are also possible, which may be sufficient even without a predicted trajectory for a motor vehicle. For example, the theoretical trajectory can be modified, for instance, by virtually moving the target position, and the actual value can be compared with the corresponding value of the parameter based on the modified theoretical trajectory.

[0014] The determination of the at least one prediction bias can also be based, for example, on assumptions about the driver's expected behavior. For instance, it can be assumed that the driver steers during parking maneuvers when the vehicle is parallel to the ground. In lateral or angled parking, it can be assumed, for example, that the driver adjusts the steering angle to approximately zero degrees at a specific point in time during parking maneuvers. Thus, the at least one prediction bias can be determined with greater reliability.

[0015] Since the generation of the tactile signal is based on at least one predicted deviation rather than, for example, the current adjustment deviation of the actual value, it is important to consider that there is typically a certain time elapsed between the output of the tactile signal and the driver's implementation of the steering suggestion, during which the vehicle continues to move. This is attributed to the driver's reaction time and the time required for the driver to execute the steering suggestion. By using the predicted deviation, the steering suggestion in the form of a tactile signal is output at an earlier time point, when implementing the steering suggestion is not yet immediately needed. However, by compensating for this time difference through the aforementioned delay, the steering suggestion can actually be implemented at the optimal time point. This improves the outcome of parking maneuvers, i.e., improves the reliable arrival at the target position. The somewhat preventative output of the steering suggestion also increases the probability that the driver will more or less accurately follow the optimal trajectory. Consequently, if less frequent or less intense tactile signals are needed, this also improves driver comfort.

[0016] Tactile signals can be generated here through the steering system of a motor vehicle, especially through the steering torque applied to the steering system, or through tactile actuators, for example, through tactile actuators used to generate vibrations as tactile signals at the steering wheel or in the driver's seat of the motor vehicle.

[0017] According to at least one implementation, the vehicle speed of the motor vehicle is determined at least approximately at the point in time when the actual value is determined. The at least one prediction deviation is determined based on the vehicle speed.

[0018] Vehicle speed can be determined, for example, by using the vehicle's speed sensor or by using the wheel position along with the rim circumference.

[0019] Therefore, it can be considered that the greater the vehicle speed, the greater the deviation between the actual trajectory and the theoretical trajectory, or the greater the deviation between the vehicle's actual final position and the target position. Correspondingly, this can further improve the results of parking maneuvers.

[0020] According to at least one embodiment, in order to generate tactile signals, additional steering torque is generated in the steering system of the motor vehicle, particularly by means of the actuator unit of the motor vehicle.

[0021] The actuator unit can intervene at any point in the steering system to generate additional steering torque. This additional steering torque can then be felt by the driver as a tactile signal at the steering wheel. The direction of the additional steering torque is specifically chosen such that if the driver follows the steering recommendation in response to the tactile signal, the direction of the additional steering torque corresponds to a reduction in the at least one predicted deviation.

[0022] Additional steering torque can be understood here as the steering torque other than that adjusted by the driver to steer the vehicle. Therefore, if the driver does not apply steering torque, the additional steering torque can be the only non-zero steering torque in the steering system. In particular, depending on the possible reset torque, the value of the additional steering torque can be higher or lower than the threshold for active steering intervention. In any case, the additional steering torque is not necessarily sufficient for fully automatic lateral control of the vehicle.

[0023] By using additional steering torque as a tactile signal, particularly intuitive steering suggestions can be provided to the driver. In particular, steering torque can be continuously output without distracting the driver from manual steering tasks. This further improves the reliability of parking maneuvers or parking assistance.

[0024] According to at least one embodiment, in particular, a first reference trajectory is determined as the at least one reference trajectory based on sensor data using a computing unit, wherein the theoretical trajectory includes a target position for parking maneuvering of the motor vehicle.

[0025] The theoretical trajectory specifically corresponds to the optimal path used to move the vehicle from its starting position to its target position. In particular, the theoretical trajectory connects the actual position of the vehicle at the start or initial time of the parking maneuver with the target position.

[0026] The theoretical trajectory can be determined reliably, thus this implementation method leads to reliable results in parking maneuvering.

[0027] According to at least one implementation method, particularly by means of a computing unit, the trajectory of a motor vehicle is predicted based on actual values. The computing unit determines the deviation between parameters based on the theoretical trajectory and parameters based on the predicted trajectory as a first prediction deviation of the at least one prediction deviation.

[0028] In particular, the prediction of the trajectory can be based on the actual value and other historical actual values, that is, actual values ​​determined before the time point when the actual value was determined.

[0029] To determine the deviation, especially in terms of parameters, points on the theoretical trajectory are compared with points on the predicted trajectory. This allows for the prediction of where the vehicle might be located on the predicted trajectory at a specific time point and where it should be located according to the theoretical trajectory at that time point. The parameters corresponding to these two locations on either the theoretical or predicted trajectory are then compared.

[0030] This introduces a time offset for generating tactile signals, which gives the driver enough time to react to steering suggestions generated by the tactile signals.

[0031] According to at least one embodiment, a tactile signal is generated based on a first prediction deviation.

[0032] According to at least one embodiment, particularly by means of a computing unit based on sensor data, a second reference trajectory is determined as the at least one reference trajectory, wherein the modified theoretical trajectory includes a target position for the vehicle modified relative to the target position. The modified target position is particularly related to the actual value of a parameter or to a target position predicted based on the predicted trajectory.

[0033] Therefore, the modified theoretical trajectory can be understood as a shifted version or variant of the theoretical trajectory. For example, the modified theoretical trajectory may correspond to a modified theoretical trajectory for a target position used by a motor vehicle. In other words, the target position is modified to obtain a virtual theoretical trajectory that the motor vehicle should ideally follow when it is to be guided to the modified target position. However, the modified target position is not the actual target position here. Thus, a time offset can also be effectively introduced, so that the adjustment deviation of the actual value used to generate the tactile signal is replaced by a predicted or expected adjustment deviation. Therefore, the modification of the target position or theoretical trajectory can be understood as a prediction. The modification of the target position or theoretical trajectory is particularly carried out in such a way that the driver's prescribed response to the corresponding tactile signal is at least partially compensated for the modification.

[0034] The reset torque of the steering system can also be taken into account by modifying the theoretical trajectory or target position. This reset torque can be adjusted, for example, when the driver turns the vehicle while it is stationary.

[0035] According to at least one implementation, particularly by means of a computing unit, the deviation between the parameters of the modified theoretical trajectory and the actual values ​​is determined as a second prediction deviation of the at least one prediction deviation.

[0036] According to at least one embodiment, a tactile signal is generated based on a second prediction deviation.

[0037] According to at least one embodiment, the deviation between the parameters of the modified theoretical trajectory and the parameters of the predicted trajectory is determined as an additional prediction deviation of the at least one prediction deviation.

[0038] According to at least one embodiment, a tactile signal is generated based on the additional prediction deviation.

[0039] In such an implementation, modifications to the theoretical trajectory or target location can therefore be correlated with predictions of the vehicle's trajectory to obtain more reliable results.

[0040] According to at least one implementation, particularly by means of a computing unit, a modified actual value is determined based on the actual value, and the deviation between the parameters based on the theoretical trajectory and the modified actual value is determined as a third prediction deviation of the at least one prediction deviation.

[0041] Similar to how the target location is moved or modified to produce a modified theoretical trajectory, as mentioned above, the actual value or predicted trajectory can also be modified in the opposite way.

[0042] According to at least one embodiment, a tactile signal is generated based on a third prediction deviation.

[0043] According to at least one embodiment, a tactile signal is generated only when one of the at least one prediction deviations is greater than or equal to a preset minimum deviation.

[0044] In other words, if no predicted deviation is greater than or equal to the minimum deviation, no tactile signal is generated. This introduces a dead zone where deviations do not lead to steering suggestions, as corresponding deviations within the parking maneuvering range result in an acceptable deviation from the target position. This further improves driver comfort without compromising the reliability of the overall parking process.

[0045] According to the improved scheme, a driver assistance system for a motor vehicle is also described, which assists in parking during manual steering parking operations. The driver assistance system includes an environmental sensor system configured to generate sensor data representing the environment of the motor vehicle. The driver assistance system includes a computing unit configured to determine at least one reference trajectory for the motor vehicle based on the sensor data. The driver assistance system also includes a sensor system configured to determine actual values ​​of parameters of the motor vehicle that can vary during the parking operation at a given time point. The computing unit is configured to determine at least one predicted deviation between the parameters and theoretical values ​​based on the at least one reference trajectory and the actual values. The driver assistance system includes an actuator unit configured to automatically generate tactile signals as steering suggestions for the driver of the motor vehicle, specifically controlled by the computing unit, based on the at least one predicted deviation.

[0046] Other embodiments of the driver assistance system according to the improved scheme are derived directly from different design approaches of the method according to the improved scheme, and vice versa. In particular, the driver assistance system according to the improved scheme can be configured or programmed to perform the method according to the improved scheme, or the driver assistance system performs such a method.

[0047] According to the improved design, a motor vehicle with a driver assistance system according to the improved design is also described.

[0048] The present invention also includes an extension of the motor vehicle according to the invention, which has features already described in conjunction with the extension of the method according to the invention. For this reason, the corresponding extension of the motor vehicle according to the invention will not be described again here.

[0049] The present invention also includes combinations of features of the described embodiments. Attached Figure Description

[0050] Embodiments of the present invention are described below. Wherein:

[0051] Figure 1 A schematic diagram of a motor vehicle having an exemplary embodiment of a driver assistance system according to an improved scheme is shown;

[0052] Figures 2a) to 2e) This illustrates different scenarios during assisted parking of a motor vehicle;

[0053] Figures 3a) to 3d) Different adjustment characteristics for generating steering recommendations are shown according to an exemplary implementation of the method based on the improved scheme; and

[0054] Figures 4a) to 4b) Other exemplary cases are shown in the context of assisted parking of a motor vehicle. Detailed Implementation

[0055] The embodiments described below are preferred embodiments of the invention. In these embodiments, the described components each represent individual features of the invention that should be considered independently of each other, and these features also independently improve the invention and can therefore be considered individually or in combinations different from those shown as part of the invention. Furthermore, the described embodiments can also be supplemented by other features among the features already described in the invention.

[0056] Components with the same function are given the same reference numerals in the figure.

[0057] exist Figure 1 The diagram schematically illustrates a motor vehicle 1 having an exemplary embodiment of a driver assistance system 2 according to an improved scheme. The driver assistance system 2 includes an environmental sensor system 3, which may include, for example, one or more ultrasonic sensor systems. The driver assistance system 2 also includes a computing unit 4 connected to the environmental sensor system 3, which may be, for example, part of or include an electronic controller of the motor vehicle 1. Furthermore, the driver assistance system 2 includes a sensor system 5 that can determine parameters of the motor vehicle 1 that vary during driving, particularly during parking maneuvers. These parameters may correspond, for example, to the position or orientation of the motor vehicle 1, the trajectory curvature of the motor vehicle 1, or the steering angle. Therefore, the sensor system 5 may include, for example, a steering angle sensor and / or one or more sensors for (e.g., programmed) determining the position or orientation of the motor vehicle 1, such as one or more wheel speed sensors, speed sensors, and / or acceleration sensors. The sensor system 5 may also include a receiver for a Global Navigation Satellite System (GNSS), such as a GPS receiver.

[0058] Furthermore, the driver assistance system 2 has an actuator unit 6 that can output tactile signals to the driver of the vehicle 1. The actuator unit 6 may be arranged, for example, in or at the steering system of the vehicle 1, to generate additional steering torque as a tactile signal, which is perceptible to the driver, for example, at the steering wheel of the vehicle 1.

[0059] The driver assistance system 2 can assist the driver when the vehicle 1 is manually parked into a parking space. For example, if the driver passes a parking space, the parking space can be identified and, if necessary, characterized by an environmental sensor system 3. The driver can, for example, stop the vehicle 1 and engage reverse gear to park laterally or parallel to the parking space.

[0060] The calculation unit 4 determines a reference trajectory based on sensor data generated by the environmental sensor system 3. In different embodiments, the reference trajectory may be a theoretical trajectory 7 for the vehicle 1, in order to guide the vehicle 1 to the target location 8 in the most optimal manner (see [link]). Figures 2a) to 2e) and Figures 4a) to 4b) ).

[0061] Sensor system 5 is configured, for example, to continuously or repeatedly determine the actual values ​​of variable parameters during parking maneuvers. Calculation unit 4 can predict the deviation between the parameters and theoretical values ​​based on a reference trajectory and the actual values. Based on the predicted deviation, actuator unit 6 can generate tactile signals as steering suggestions for the driver.

[0062] If the reference trajectory is the theoretical trajectory 7, then the calculation unit 4 can determine the predicted trajectory 7' for vehicle 1, for example, based on the actual values ​​of the parameters and, if possible, other historical actual values ​​of the parameters. Figures 2a) to 2e) As illustrated in the image. Prediction bias corresponds, for example, to the deviation in position or trajectory curvature between the theoretical trajectory 7 and the predicted trajectory 7' at a later time point.

[0063] Therefore, the tactile signals used for steering recommendations are not generated based on the current accommodation bias, but rather on the accommodation bias predicted from the theoretical value.

[0064] exist Figures 2a) to 2e) Images a) through e) illustrate different scenarios in which vehicle 1 is guided to the target position 8 into a parking space between two other vehicles 1', 1''. Images a) and b) show parking maneuvers for rear-facing lateral parking, while images c) through e) show parking maneuvers for rear-facing parallel parking between the other vehicles 1', 1''.

[0065] The predicted trajectory 7' corresponds to different deviations between the driver's behavior and the optimal behavior for following the theoretical trajectory 7. In the case of image a), the driver of vehicle 1 yaws too little or too late, while in the case of image b), the driver yaws too early or too much. Accordingly, actuator unit 6 generates a rightward steering torque in the case of image a) and a leftward steering torque in the case of image b), in order to induce the driver to yaw more or less strongly.

[0066] In picture c), the driver of vehicle 1 veers too early or too much; in picture d), the driver veers too late or too little; and in picture e), the driver veers too early or returns to the starting position too late.

[0067] Different strategies for adjusting parameters are illustrated in Figures 3a) to 3d) Images a) through d) are shown. Here, the prediction bias is plotted on the horizontal axis, and the intensity of the tactile signal, particularly the intensity of the additional steering torque, is plotted on the vertical axis. For example, the additional steering torque can be limited by a minimum or maximum additional steering torque, such as... Figures 3a) to 3d) As shown in the various pictures.

[0068] In Figure a), the intensity of the additional steering torque varies linearly between the maximum and minimum additional steering torque. In Figure b), a dead zone is additionally set for small prediction biases, in which no additional steering torque is generated. For example, outside the dead zone, the relationship is also linear. Figures c) and d) schematically illustrate the superlinear relationship between prediction bias and additional steering torque.

[0069] exist Figures 4a) to 4b) The diagrams schematically illustrate other scenarios in the parking of vehicle 1. Image a) shows the process for lateral parking, while image b) shows the process for parallel parking. In addition to the corresponding theoretical trajectory 7, two predicted trajectories 7' are also shown, where the steering is too early and / or too strong, or too late and / or too weak. This results in corresponding deviations in the target position 8, as indicated by the arrows.

[0070] According to the improved design, manual steering can therefore be performed with greater reliability and an improved subjective steering feel for the driver. In particular, the driver can thus be guided so that not only the timing of the driver's steering action but also the intensity of that steering action results in the best possible trajectory and therefore the best possible parking position and parking orientation.

[0071] The driver should follow the optimal trajectory as precisely as possible. This requires time for the driver to perceive and understand the steering suggestion and until the expected response is implemented. To ensure reaching the optimal stopping position, the driver's reaction time can also be considered by predicting deviations in an improved scheme. Therefore, steering suggestions can be given in time, before they should be implemented. The timing of the steering suggestion can also be designed in relation to vehicle speed.

[0072] According to the improved scheme, the steering suggestion is not determined based on the actual adjustment deviation, but rather on the predicted adjustment deviation, which can be determined between theoretical and actual values ​​of steering angle, vehicle position, vehicle orientation, etc. In different implementation schemes, fixed values ​​for time and / or location offsets to the target position can also be preset independently of the predicted trajectory.

[0073] In different implementations, when a deviation from the theoretical trajectory exists, a minimum torque can be applied independently of the magnitude of the deviation. This minimum torque is perceptible so that a reset torque can also be used or compensated for, in particular. In different designs, the target position can be dynamically moved to correct the deviation based on whether the driver has already turned while stationary.

[0074] List of reference numerals

[0075] 1,1',1'' Motor vehicles

[0076] 2 Driver assistance systems

[0077] 3. Environmental Sensor System

[0078] 4. Calculation Unit

[0079] 5. Sensor System

[0080] 6 Actuator Units

[0081] 7,7' trajectory

[0082] 8. Target location.

Claims

1. A method for parking assistance for a motor vehicle (1) in manual steering parking operation, wherein - Sensor data representing the environment of the vehicle (1) is generated using the environmental sensor system (3) of the vehicle (1); - Using the computing unit (4) of the motor vehicle (1), at least one reference trajectory for the motor vehicle (1) is determined based on the sensor data; - The calculation unit (4) uses the sensor system (5) of the motor vehicle (1) to determine the actual value of the variable parameters of the motor vehicle (1) during the parking operation at a specific point in time during the parking operation; Its features are, - Using the calculation unit (4), at least one prediction deviation between the parameter and the theoretical value is determined based on the at least one reference trajectory and the actual value, wherein the prediction deviation of the parameter is the deviation between the estimated or predicted parameter and the theoretical value at a time point after the time point where the actual value is determined; as well as - Automatically generate tactile signals as steering suggestions for the driver of the motor vehicle (1) based on the at least one prediction deviation.

2. The method according to claim 1, characterized in that, Additional steering torque is generated in the steering system of the motor vehicle (1) to produce a tactile signal.

3. The method according to any one of the preceding claims, characterized in that, The theoretical trajectory (7) for the motor vehicle (1) is determined as the first reference trajectory of the at least one reference trajectory, wherein the theoretical trajectory (7) includes the target position (8) for parking maneuvering of the motor vehicle (1).

4. The method according to claim 3, characterized in that, - Predict the trajectory of the motor vehicle (1) based on the actual value; and - The first prediction deviation is determined as the deviation between the parameters of the theoretical trajectory (7) and the parameters of the predicted trajectory.

5. The method according to claim 3, characterized in that, - The modified theoretical trajectory used for the motor vehicle (1) is determined as a second reference trajectory for the at least one reference trajectory; and - The modified theoretical trajectory includes the target position for the motor vehicle (1) modified relative to the target position (8).

6. The method according to claim 5, characterized in that, The second prediction deviation is determined as the deviation between the parameters of the modified theoretical trajectory and the actual values.

7. The method according to claim 3, characterized in that, - Determine the modified actual value based on the actual value; and - The deviation between the parameters of the theoretical trajectory (7) and the modified actual values ​​is determined as an additional prediction deviation of the at least one prediction deviation.

8. The method according to claim 1 or 2, characterized in that, A tactile signal is generated only when one of the at least one prediction deviations is greater than or equal to a preset minimum deviation.

9. A driver assistance system for a motor vehicle (1), the driver assistance system being used for parking assistance during manual steering parking operations, the driver assistance system (2) comprising: - An environmental sensor system (3) configured to generate sensor data representing the environment of the motor vehicle (1); - A calculation unit (4), configured to determine at least one reference trajectory for the motor vehicle (1) based on the sensor data; and - Sensor system (5), the sensor system being configured to determine the actual values ​​of parameters of the motor vehicle (1) that can vary during parking maneuvers at a point in time during parking maneuvers; Its features are, - The calculation unit (4) is configured to determine at least one prediction deviation between the parameter and the theoretical value based on the at least one reference trajectory and based on the actual value, wherein the prediction deviation of the parameter is the deviation between the estimated or predicted parameter and the theoretical value at a time point after the time point in which the actual value is determined; and - The driver assistance system (2) has an actuator unit (6) configured to automatically generate tactile signals as steering suggestions for the driver of the motor vehicle (1) based on the at least one prediction deviation.

10. A motor vehicle having a driver assistance system (2) according to claim 9.

Images