Method for operating a driver assistance system and driver assistance system
By integrating real-time and historical data to correct measurement errors, the method improves the accuracy and safety of driver assistance systems by ensuring accurate driving strategies.
Patent Information
- Application Number
- DE102024201053
- Authority / Receiving Office
- DE · DE
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-02-06
- Publication Date
- 2025-08-07
AI Technical Summary
Existing driver assistance systems face inaccuracies due to measurement errors in object detection, leading to potential safety issues such as unnecessary emergency braking.
A method that combines real-time and historical object data to calculate a dynamic driving strategy, correcting measurement errors by comparing and prioritizing data based on selection parameters like speed and position, ensuring accurate and reliable operation.
Enhances the accuracy and reliability of driver assistance systems by reducing the impact of measurement errors, improving vehicle safety through more appropriate driving strategies.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a method for operating a driver assistance system according to claim 1. Furthermore, the invention relates to a driver assistance system. State of the art
[0002] DE 10 2019 113 345 A1 describes a system for object detection and tracking in vehicles. It involves capturing radar measurements at different times, organizing these measurements into chronologically ordered clusters, and forming a sequence cluster from this data. The radar data is linked to road and lane information, and the movement of the detected objects within the recorded lanes is tracked using a restricted filter by comparing the clusters with a road topology map, and conflicting radar measurements are removed. Disclosure of the invention
[0003] According to the present invention, a method for operating a driver assistance system of a vehicle is proposed, having the features of claim 1. This allows the driver assistance system to be operated more accurately and reliably. Furthermore, the vehicle safety of the vehicle can be increased. The reliability of the acquired real-time object data can be increased. The real-time object data can be checked for plausibility. Measurement errors occurring during the acquisition of the object data can be detected and compensated. The occurrence of apparent speeds due to measurement errors during the acquisition of the object data can be reduced.
[0004] The vehicle can be a motor vehicle, a two-wheeled vehicle, or a truck. The vehicle can be a motor-driven vehicle. The vehicle can be a mobile robot. The vehicle can be semi-autonomous or autonomous.
[0005] The environmental object can be a vehicle, a living being or at least one object, in particular an article, a facility, a device.
[0006] The object data can include a dimension as a parameter. The kinematic data can include a position, a speed, an orientation, a rotational speed, an acceleration, and / or a rotational acceleration. The kinematic data of the environment object can describe the movement of the environment object.
[0007] Acquiring the object data may include measuring and processing the object data. Acquiring the real-time object data may include measuring and processing real-time object data. Acquiring may include environmental perception, in particular, object modeling of at least one object model of the environmental object. The object data and / or real-time object data may be measured by a vehicle sensor system. The vehicle sensor system may include at least one radar sensor, an ultrasonic sensor, a LIDAR sensor, and / or a video camera for environmental detection.
[0008] The historical object data can be real-time object data recorded prior to the current point in time. The historical object data can be stored at predefined points in time. The time interval between the points in time of the historical object data can be constant or variable.
[0009] The traveled object trajectory is the spatial path the surrounding object has traveled so far. The traveled object trajectory can also be calculated based on map data from a map of the vehicle's surroundings. In addition to the position of the surrounding object, the traveled object trajectory can also contain information about the speed, orientation, pose, angular velocity, acceleration, and / or angular acceleration of the surrounding object at the assigned position.
[0010] The current estimated object data can specify an estimated position, velocity, orientation, pose, rotational velocity, acceleration, and / or angular acceleration of the surrounding object as parameters. The current estimated object data can be calculated by extrapolating the calculated object trajectory. The current estimated object data can further be calculated depending on the real-time object data.
[0011] If the comparison reveals a deviation between the real-time object data and the estimated object data or if the deviation exceeds a specified threshold, the real-time object data can be replaced by the estimated object data or the plausibility of the real-time object data can be reduced for further use of the real-time object data (e.g. for emergency braking).
[0012] Calculating the driving strategy can include planning the driving strategy for at least one point in time following the current point in time. The vehicle's driving strategy can be a dynamic driving strategy. The dynamic driving strategy can include emergency braking. The reason and / or severity of the driving strategy, in particular the emergency braking, can be calculated based on the comparison. The driving strategy can specify a planned position, speed, orientation, pose, acceleration, and / or angular acceleration of the vehicle following the current point in time.
[0013] The dynamic driving strategy can be assigned to the driver assistance system. The dynamic driving strategy can be applied during semi-autonomous or autonomous driving of the vehicle. The driver assistance system can control semi-autonomous or autonomous driving of the vehicle.
[0014] The method for operating the driver assistance system may be a computer-implemented method.
[0015] In a preferred embodiment of the invention, it is advantageous if the real-time object data is modified depending on the comparison and output as output object data. The output object data can have at least one modified parameter compared to the real-time object data. The parameter can be a position, a speed, an orientation, a pose, an acceleration, and / or a rotational acceleration of the surrounding object.
[0016] In a preferred embodiment of the invention, the calculation of the dynamic driving strategy is performed depending on the output object data. This allows the dynamic driving strategy to be calculated more accurately and appropriately.
[0017] In an advantageous embodiment of the invention, the historical object data is accumulated over a period preceding the current time. The historical object data can be formed from real-time object data of the period. For this purpose, the real-time object data can be stored at the times within the period.
[0018] In a preferred embodiment of the invention, it is advantageous if the comparison is preceded by a selection of at least one parameter of the real-time object data as a selection parameter, and the comparison is carried out between the selection parameter and at least one parameter determined from the calculated object trajectory. The selection parameter can be a position, a speed, in particular a lateral or longitudinal speed of the surrounding object relative to the vehicle, a rotational speed, a pose, an acceleration, and / or a rotational acceleration of the surrounding object. The parameter that is considered in comparison with the selection parameter can be calculated from the calculated object trajectory.
[0019] In a specific embodiment of the invention, it is advantageous if the selection parameter is selected dynamically while the vehicle is driving. The selection parameter can be a position, speed, rotational speed, pose, acceleration, or rotational acceleration of the surrounding object.
[0020] In a preferred embodiment of the invention, the selection parameter is selected depending on the current operating state, the current driving situation, and / or traffic situation of the vehicle. The driving situation can refer to a driving operation of the vehicle. The traffic situation can refer to an interaction with the vehicle's surroundings, in particular with the surrounding object.
[0021] For example, if the surrounding object is moving toward the vehicle from the side, the lateral velocity of the surrounding object can be specified as a selection parameter. If the surrounding object is moving toward the vehicle from the front, the longitudinal velocity of the surrounding object can be specified as a selection parameter.
[0022] A preferred embodiment of the invention is advantageous in that, during the comparison, the real-time object data is evaluated based on the estimated object data, and the driving strategy is calculated based on the evaluation. The evaluation of the real-time object data can include an evaluation of the selection parameter.
[0023] The evaluation may include prioritizing the real-time object data, in particular the selection parameter, over the estimated object data, in particular over the parameter of the estimated object data. The evaluation may be performed to give priority to the real-time object data over the estimated object data, or vice versa. The evaluation may be a plausibility check of the real-time object data. The evaluation may result in such low plausibility of the real-time object data that it is considered too unreliable, inaccurate, and / or unsafe for the operation of the driver assistance system. As a result, a vehicle function, such as emergency braking, may not be initiated based on the real-time data.
[0024] In an advantageous embodiment of the invention, the evaluation is dependent on the vehicle's operating state, driving situation, and / or traffic situation. The real-time object data can be given priority over the estimated object data, for example, in the case of active adaptive cruise control. The real-time object data can be placed downstream of the estimated object data, for example, to control an emergency braking system.
[0025] According to the present invention, a driver assistance system with the features of claim 10 is further proposed. The driver assistance system can execute lane keeping control, emergency braking, and / or speed or distance control. The calculated driving strategy can be associated with lane keeping control, emergency braking, and / or speed or distance control.
[0026] Further advantages and advantageous embodiments of the invention emerge from the description of the figures and the illustrations. Character description
[0027] The invention is described in detail below with reference to the figures. They show in detail: Fig. 1: A method for operating a driver assistance system in a specific embodiment of the invention. Fig. 2: A representation of captured object data of an environmental object. Fig. 3: Another representation of captured object data of an environmental object.
[0028] Fig. 1 shows a method for operating a driver assistance system in a specific embodiment of the invention. The method for operating 10 a driver assistance system 11 of a vehicle comprises capturing 12 object data 14 of an environmental object 16 in a vehicle environment of the vehicle. The environmental object 16 can, for example, be another vehicle 18. The object data 14 comprise real-time object data 20, i.e., object data 14 of the environmental object 16 captured at the current time and historical object data 22 of the environmental object 16 preceding this data. The real-time object data 20 can, for example, be measured and processed by a vehicle sensor system, in particular with at least one radar sensor or a video camera. The historical object data 22 can be accumulated over a period T preceding the current time ta from the real-time object data 20 captured and stored at the times in the period.
[0029] The historical object data 22 are used to calculate 24 a traveled object trajectory 26 of the surrounding object 16. The traveled object trajectory 26 is the path that the surrounding object 16 has spatially traveled so far. In addition to the position of the surrounding object 16, the traveled object trajectory 26 can also contain information about the speed and / or acceleration of the surrounding object 16 at the respective points in time.
[0030] Subsequently, a calculation 28 of current estimated object data 30 is carried out depending on the calculated object trajectory 26. The current estimated object data 30 can be calculated by extrapolating the calculated object trajectory 26.
[0031] The estimated object data 30 are compared with the real-time object data 20 by a comparison 32. The real-time object data 20 are modified depending on the comparison 32, in particular their plausibility is changed, for example, downgraded, and output as output object data 34. The comparison 32 is preceded by a selection 36 of at least one parameter of the real-time object data 20 as a selection parameter 38, and the comparison 32 is performed between the selection parameter 38 and at least one parameter determined from the calculated object trajectory 26, preferably a parameter 40 of the estimated object data 30. The selection parameter 38 is, for example, fixedly specified or selected dynamically during vehicle operation, for example, depending on the current driving situation and / or traffic situation of the vehicle.During the comparison 32, an evaluation 42, in particular a plausibility check, of the real-time object data 20, in particular of the selection parameter 38, is performed, and the driving strategy 44 is calculated depending on the evaluation 42. The evaluation 42 can, for example, be a prioritization 46 of the selection parameter 38 over the parameter 40 of the estimated object data 30. The output object data 34 can include the selection parameter 38 or the parameter 40 of the estimated object data 30, depending on the comparison 32.
[0032] The calculation 48 of a dynamic driving strategy 44 of the vehicle is then carried out depending on the comparison 32, specifically depending on the output object data 34.
[0033] Fig. Figure 2 shows a representation of recorded object data of an environmental object. The environmental object 16 is another vehicle 18, which is actually stationary and not moving at times t1 to t4. The actual speed of the environmental object 16 is therefore zero. The other vehicle 18 is also stationary at the current time ta.
[0034] The acquired object data includes the dimensions 50 of the environmental object 16 and the kinematic data with the parameters position 52 and velocity 54 of the environmental object 16. The acquired object data is formed from the real-time object data and the historical object data. The real-time object data indicates the acquired dimensions 50a, the acquired position 52a, and the acquired velocity 54a at the current time ta. The historical object data includes, for example, the dimensions 50, the position 52, and the velocity 54 of the environmental object 16 at times t1, t2, t3, and t4.
[0035] The detected speed 54 at time t2 is greater than zero, while the actual speed of the other vehicle 18 is zero. The same applies to times t3 and t4. Also, the detected position 52 at times t1 to t4 is different from the actual position 56.
[0036] The current detected speed 54a at time ta is greater than zero, while the actual speed of the other vehicle 18 is zero. Also, the current detected position 52a of the other vehicle 18 is different from the current actual position 56a of the other vehicle 18.
[0037] The detected speeds 54, 54a greater than zero are apparent speeds resulting from measurement errors when capturing the object data. If these measurement errors are not detected and taken into account, vehicle safety may be compromised, for example, by triggering an emergency braking of the vehicle even though the other vehicle 18 poses no threat to the vehicle.
[0038] Through the exemplary Fig. 1 described method for operating the driver assistance system, the dynamic driving strategy can better correspond to the driving situation and traffic situation of the vehicle by controlling the driver assistance system.
[0039] Fig.Figure 3 shows another representation of recorded object data of an environmental object. The environmental object 16 is another moving vehicle 18, which has an actual speed 58 at times t1 to t3. At the current time ta, the other vehicle 18 also has the actual speed 58. The actual speed 58 of the other vehicle 18 is constant over time in terms of magnitude and direction from times t1 to t4 and ta.
[0040] The detected position 52 and speed 54 of the other vehicle 18 at times t1 to t3 deviate from the actual position 56 and actual speed 58. The current detected position 52a and speed 54a also deviate from the actual position 56a and actual speed 58a of the other vehicle 18 at the current time ta.
[0041] The operating method calculates the traveled object trajectory 26 from the historical object data from times t1 to t3, and calculates the currently estimated object data based on the calculated object trajectory 26. The currently estimated object data with the current estimated position 60 and the current estimated speed 62 may deviate from the actual current position 56a and speed 58a, but may be much closer to the actual values than the currently recorded parameters. QUOTES CONTAINED IN THE DESCRIPTION
[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature
[0000] DE 10 2019 113 345 A1
[0002]
Claims
[1] Method for operating (10) a driver assistance system (11) of a vehicle, comprising the steps Acquiring (12) object data (14) comprising kinematic data of an environmental object (16) in a vehicle environment of the vehicle, wherein the object data (14) comprise real-time object data (20) and historical object data (22) preceding said real-time object data, Calculating (24) a traveled object trajectory (26) of the surrounding object (16) at least from the historical object data (22), Calculation (28) of current estimated object data (30) depending on the calculated object trajectory (26), Comparison (32) of the real-time object data (20) with the estimated object data (30), calculation (48) of a driving strategy (44) of the vehicle depending on the comparison (32). [2] Method of operation (10) according to claim 1, characterized bythat the real-time object data (20) are changed depending on the comparison (32) and output as output object data (34). [3] Method of operation (10) according to claim 2, characterized by that the calculation (48) of the dynamic driving strategy (44) is carried out depending on the output object data (34). [4] Method of operation (10) according to one of the preceding claims, characterized by that the historical object data (22) are accumulated over a period (T) preceding the current time (ta). [5] Method of operation (10) according to one of the preceding claims, characterized by that the comparison (32) is preceded by a selection (36) of at least one parameter (40) of the real-time object data (20) as a selection parameter (38) and the comparison (32) is carried out between the selection parameter (38) and at least one parameter (40) determined from the calculated object trajectory (26). [6] Method of operation (10) according to claim 5, characterized by that the selection parameter (38) is selected dynamically during driving of the vehicle. [7] Method of operation (10) according to claim 5 or 6, characterized by that the selection parameter (38) is selected depending on the current operating state, the current driving situation and / or traffic situation of the vehicle. [8] Method of operation (10) according to one of the preceding claims, characterized by that during the comparison (32) an evaluation (42) of the real-time object data (20) is carried out depending on the estimated object data (30) and the calculation (48) of the driving strategy (44) is carried out depending on the evaluation (42). [9] Method of operation (10) according to claim 8, characterized by that the rating (42) depends on an operating condition, a driving situation and / or a traffic situation of the vehicle. [10] Driver assistance system (11) for a vehicle, wherein the driver assistance system (11) is operable by an operating method (10) according to one of the preceding claims.
Citation Information
Patent Citations
SYSTEMS AND METHODS FOR USING ROAD UNDERSTANDING TO RESTRICT RADAR LANES
DE102019113345A1