Method and system for testing driver assistance systems

Abstract

The present invention relates to a computer-implemented method for testing a driver assistance system of an ego vehicle based on test drive data, the method comprising the steps of: assigning attributes to other vehicles captured in the test drive data that are particularly close to the ego vehicle, the attributes specifying the relative positions of the other vehicles to the ego vehicle at times in the test drive data and associated with relevant times; examining the test drive data for the occurrence of basic lateral movements characterized in each case by a change in the position of the ego vehicle or one of the other vehicles perpendicular to the course of the road, and basic longitudinal movements characterized in each case by a change in the leading distance and / or trailing distance of the ego vehicle or one of the other vehicles, particularly in the same lane, where the basic movements are selected from a list of predetermined basic movements, and the occurrence of a basic movement is also associated with at least one relevant time point; identifying the occurrence of predetermined scenarios based on the occurred basic movements, where the predetermined scenarios are characterized by a collection of basic movements and attributes; and analyzing the driving behavior of the driver assistance system in the identified scenarios. The present invention also relates to a corresponding system.

Description

Technical Field

[0005] ,

[0007] , , ,

[0006]

[0001] The present invention relates to a computer-implemented method for testing a driver assistance system of an ego vehicle based on test drive data.

Background Art

[0002] In both the fields of passenger cars and commercial vehicles, the spread of driver assistance systems (Advanced Driver Assistance Systems (ADAS)) has been steadily progressing. The contribution of driver assistance systems to the expansion of active traffic safety is important and helps to enhance the comfort of driving.

[0003] In the fields of passenger cars and commercial vehicles, in addition to systems that are particularly useful for safe driving, such as ABS (Anti-lock Braking System) and ESP (Electronic Stability Program), a plurality of driver assistance systems, such as garage entry assistance, adaptive cruise control, and lane assistance, are recommended. These driver assistance systems not only improve traffic safety by warning the driver of critical situations, but also initiate autonomous intervention to prevent accidents or mitigate the impact of accidents, for example, by activating an emergency braking function. In addition, functions such as automatic garage entry, automatic lane keeping, and automatic proximity control enhance the comfort of driving.

[0004] The improvement in safety and comfort by the assistance system is positively perceived by the vehicle occupants only when the assistance by the driver assistance system is safe, reliable, and as convenient as possible.

[0005] Furthermore, all driver assistance systems need to handle a given traffic scenario with the highest safety for the vehicle itself and without exposing other vehicles or other road users to danger, respectively, depending on their functions.

[0006] Therefore, it is necessary to analyze and optimize the driver assistance system or the driving behavior by the driver assistance system, respectively.

[0007] Document WO2015 / 032508 is, A step to check whether at least one driver assistance system A is activated, A step of detecting at least one vehicle parameter function suitable for characterizing the operating state of the vehicle and / or at least one environmental parameter function suitable for characterizing the environment of the vehicle, In particular, the steps include determining at least one characteristic function of a driving condition that characterizes the driving conditions of a vehicle, based at least on at least one vehicle parameter function and / or at least one environmental parameter function, The steps include determining at least one control intervention characteristic function suitable for characterizing the activity of at least one driver assistance system A, A correction function that relies on at least one control intervention characteristic function and at least one driving situation characteristic function, and includes the step of determining a correction function that is particularly suitable for characterizing the subjective perception of the activities of at least one vehicle occupant of the driver assistance system A, based on at least one vehicle parameter function and / or at least one environmental parameter function, This relates to a method for optimizing at least one driver assistance system, including the steps involved in the procedure. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] WO2015 / 032508 [Overview of the project] [Problems that the invention aims to solve]

[0009] One of the objectives of the present invention is to specify an improved method for testing driver assistance systems. In particular, an objective of the present invention is to improve the determination of driving conditions that occur during test driving within the scope of driver assistance system testing. [Means for solving the problem]

[0010] This problem is solved by teaching in the independent claims. Advantageous embodiments are claimed in the dependent claims.

[0011] A first aspect of the present invention relates to a computer-based method for testing a driver assistance system for an EGO vehicle based on test driving data, wherein the method is: A step of assigning attributes to other vehicles that are particularly close to the ego vehicle and have been incorporated into the test run data, wherein these attributes define the relative position of each of the other vehicles to the ego vehicle at a given point in time within the test run data and are associated with the relevant point in time. Steps to examine test drive data to determine whether, in each case, a basic lateral motion characterized by a change in the position of one of the ego vehicle or other vehicles perpendicular to the road course, and in each case, a basic longitudinal motion characterized by a change in the distance to the vehicle in front and / or the distance to the vehicle behind one of the ego vehicle or other vehicles, particularly in the same lane, is occurring, wherein the basic motion is selected from a predetermined list of basic motions, and the occurrence of the basic motion is also associated with at least one relevant point in time. A step of identifying the occurrence of a predetermined scenario based on the basic movements that have occurred, wherein the predetermined scenario is characterized by a set of basic movements and attributes, The step involves analyzing the driving behavior of the driver assistance system in the identified scenario, Includes the steps of the procedure.

[0012] A second aspect of the present invention relates to a system for functional testing of a driver assistance system for an ego vehicle based on test drive data, wherein the system is A means for assigning attributes to other vehicles, particularly those immediately preceding the ego vehicle and incorporated into test run data, wherein these attributes define the relative position of each of the other vehicles to the ego vehicle at a given point in time within the test run data, and are associated with the relevant point in time. A means for examining test drive data to determine whether, in each case, a basic lateral motion characterized by a change in the position of one of the ego vehicle or other vehicles perpendicular to the road course, and in each case, a basic longitudinal motion characterized by a change in the distance to the vehicle in front and / or the distance to the vehicle behind one of the ego vehicle or other vehicles, particularly in the same lane, is occurring, wherein the basic motion is selected from a predetermined list of basic motions, and the occurrence of the basic motion is also associated with at least one relevant point in time. A means for identifying the occurrence of a predetermined scenario based on the basic movements that have occurred, wherein the predetermined scenario is characterized by a set of basic movements and attributes, Means for analyzing the driving behavior of driver assistance systems in identified scenarios and It is equipped with.

[0013] Further aspects of the present invention relate to a computer program product containing instructions and a computer-readable medium storing such a computer program product, wherein the computer is prompted to perform a step of the method according to the first aspect of the present invention by executing the instructions.

[0014] Within the scope of the present invention, testing a driver assistance system is useful for analyzing or optimizing the driver assistance system or the driving behavior of a driver assistance system. This can also occur in the development process of road driving behavior or, in particular, virtual environments.

[0015] Means within the scope of the present invention may be configured as hardware and / or software, and in particular comprise a processing unit, in particular a digital processing unit, to which data or signals are connected, preferably a memory or bus system, and in particular a microprocessor unit (CPU) and / or one or more programs or program modules. The CPU may be designed to process instructions to be executed as a program stored in a storage system, to detect input signals from a data bus and / or to send output signals to the data bus. The storage system may comprise one or more different storage media, in particular optical storage media, magnetic storage media, solid-state storage media and / or other non-volatile storage media. A program may be provided so that the CPU can embody or perform such a method so that it can perform steps of the method described herein to analyze a vehicle under test in particular.

[0016] Scenarios within the scope of the present invention are preferably formed by a temporal sequence of spatial, particularly static, scenes. Thereafter, the spatial scenes preferably indicate the spatial arrangement of at least one other object relative to the ego vehicle, such as a group of stationary objects, such as road users or lane markings. Scenarios particularly include driving conditions in which a vehicle known as the ego vehicle, for example, equipped with a driver assistance system that autonomously performs at least one vehicle function of the ego vehicle, is at least partially controlled by the driver assistance system.

[0017] Within the scope of the present invention, a lane or traffic lane is preferably a paved road, and more specifically, a traffic lane on the road surface intended to proceed in a specified direction. Preferably, the lane or traffic lane is marked.

[0018] The basic motions within the scope of the present invention are preferably basic lateral motion, basic longitudinal motion, and / or basic cornering motion.

[0019] The basic lateral movement within the scope of the present invention is preferably a driving movement lateral to the course of the ego vehicle's travel path.

[0020] The longitudinal movement within the scope of the present invention is preferably at least substantially a driving movement in the direction of the ego vehicle's travel path.

[0021] The basic cornering movement within the scope of the present invention is preferably a driving movement in which the trajectory of the ego vehicle represents a curve.

[0022] The test drive data within the scope of the present invention preferably relates to values, and in particular to a data set of parameters characterizing the environment and / or the operation of the ego vehicle during the test drive.

[0023] The vehicle within the scope of the present invention is preferably a road user, and thus, in particular, an object moving in traffic.

[0024] 〔 The driving behavior within the scope of the present invention is preferably characterized by the driving characteristics of the driver assistance system. In particular, the driving behavior is characterized by the action of the driver assistance system in its environment and its reaction to that environment.

[0025] The present invention is based on a method for performing a scenario-based evaluation for verifying and validating the functions of a driver assistance system. In such a scenario-based evaluation, in a specific scenario, the driving behavior of the driver assistance system is observed, analyzed, and / or evaluated.

[0026] The teachings of the present invention are realized through test drive data of an ego vehicle, which is preferably captured and composed during actual driving operations and then retrieved for basic motion in a scenario. Data fields of the test drive data corresponding to a predetermined scenario related to the driver assistance system being tested are analyzed. The method of the present invention enables particularly reliable identification of the relevant data fields for each driver assistance system being tested. This results in particularly high-quality test results. Moreover, the set of test drive data from the vehicle can be repeatedly used to test various versions and / or other driver assistance systems. This makes it possible to significantly reduce the number of actual or virtual test drives required to generate the test drive data. In particular, with respect to actual test drive data, the distance typically driven by an actual driver required to generate such test drive data can be significantly reduced.

[0027] Furthermore, the method of the present invention provides test engineers with a high degree of flexibility when evaluating test run data with respect to a specific function. Specifically, test engineers can define an unlimited number of different scenarios from which test run data can be retrieved. This makes it possible to generate the optimal scenario for testing a specific function of the driver assistance system. In addition, from a series of test runs, the optimal test run data for analyzing the driving behavior of each driver assistance system can be identified.

[0028] In one advantageous embodiment of this method, test run data is retrieved solely for the attributes and / or basic motion contained within a given scenario.

[0029] In this embodiment, only data fields that may be relevant to the test run data are included in the search, so the inspection of the test run data can be significantly reduced in terms of computational power and / or computation time.

[0030] In a more advantageous embodiment of this method, test runs are performed on a test bench using test run data to analyze the driving behavior of the driver assistance system in identified scenarios. The test bench is preferably a vehicle test bench, a vehicle in the loop test bench, a hardware in the loop test bench, or a software in the loop test bench.

[0031] This embodiment can achieve particularly high quality in the analysis, evaluation, and / or optimization of the driving behavior of the driver assistance system.

[0032] In a more advantageous embodiment, test run data is checked for the occurrence of basic motions using a machine learning-trained model of basic motions. Specifically, when checking for basic motions, patterns are used to recognize basic motions in the test run data, and these patterns are generated by machine learning based on test run data that has already been classified with respect to motion. Preferably, the test run data is therefor to be classified manually and then imported into a machine learning algorithm, in particular an artificial neural network.

[0033] This embodiment has the advantage that the basic movements themselves are trained within the machine learning model process, rather than being scenarios. This provides a high degree of flexibility in defining new scenarios, as new scenarios can be compiled modularly from individual patterns or from models of the basic movements. In principle, customized scenarios can thus be compiled for their respective applications.

[0034] In a more advantageous embodiment of this method, the list includes at least one of the groups of basic lateral movements, such as changing lanes to the left, changing lanes to the right, driving within a lane, driving outside a lane, changing lanes to the right, and changing lanes to the left.

[0035] In a more advantageous embodiment of this method, the list includes at least one of a group of basic longitudinal movements, such as initial starting, increasing the distance, decreasing the distance, following a vehicle, driving in an empty lane, and stopping.

[0036] In a more advantageous embodiment of this method, the test drive data is further examined to determine whether a basic cornering motion is occurring, and the basic cornering motion is selected from a list that includes at least one of the basic cornering motion groups, such as straight driving without curvature, cornering with increasing absolute curvature, exit cornering with decreasing absolute curvature, cornering with constant curvature, left turn, right turn, and driving through a roundabout.

[0037] Including cornering motion in the basic motion allows for a more differentiated classification of test run data.

[0038] In a more advantageous embodiment of this method, the attributes indicate to the ego vehicle whether another vehicle is in the same lane, or in the right or left lane, and whether, in terms of the road course, this other vehicle is ahead of, behind, or alongside the ego vehicle. This allows other road users to be clearly identified.

[0039] In a more advantageous embodiment of this method, the attributes further indicate which vehicle in the lane is this other vehicle relative to the ego vehicle.

[0040] In a more advantageous embodiment of this method, the attribute further indicates the direction of travel of this other vehicle relative to the direction of travel of the ego vehicle.

[0041] In a more advantageous embodiment of this method, attributes are assigned only to distances defined within the measurement range of the sensor used to determine the attributes of the ego vehicle, without relying on the distance of this other vehicle to the ego vehicle. As a result, the information contained in the test drive data is reduced to information that is actually relevant to defining basic driving operations.

[0042] This can lead to faster data processing and / or require less computational power.

[0043] In a more advantageous embodiment of this method, test run data is generated based on actual test run data, and the lanes of the ego vehicle and this other vehicle are preferably determined by an intelligent camera mounted on the ego vehicle.

[0044] In a more advantageous embodiment of this method, the known locations of land landmarks on a high-resolution map captured by a reference system, specifically an intelligent camera, are further used to determine the lanes of the ego vehicle and this other vehicle.

[0045] Using intelligent cameras enables particularly segmented analysis of test run data.

[0046] In a more advantageous embodiment of this method, test run data is generated based on actual test runs, and the relative position of this other vehicle to the Ego vehicle is determined in each case preferably by intelligent cameras, lidars, and / or radar mounted on the Ego vehicle.

[0047] Further features and advantages of the present invention will become apparent from the following description with reference to the figures, which are at least partially schematic. [Brief explanation of the drawing]

[0048] [Figure 1a] This is a diagram of the Ego vehicle during test runs. [Figure 1b] This is a diagram of an exemplary embodiment of a system for testing a driver assistance system. [Figure 2] This is a flowchart illustrating an exemplary embodiment of a method for testing a driver assistance system. [Figure 3] This is a diagram that shows the attributes of other vehicles. [Figure 4] This diagram illustrates the dynamic changes in the attributes of other vehicles. [Figure 5] Figure 5a is a diagram showing the temporal sequence of overtaking movements by the Ego vehicle. Figure 5b is a graphical representation of the overtaking movement in Figure 5a. [Modes for carrying out the invention]

[0049] Figure 1a shows vehicle 2 undergoing test operation on road 5.

[0050] During the test run, Vehicle 2 collects test run data 6 as an ego vehicle acting as a reference in traffic conditions. The ego vehicle 2 preferably has multiple sensors for recording traffic conditions and the environment around the vehicle. Figure 1a shows, as just one example, an ego vehicle 2 having a camera 4, which is an intelligent camera. Preferably, such a camera 4 has a 360° field of view to monitor the entire environment around the ego vehicle 2. Further possible sensors include radar, lidar, ultrasonic, etc. The intelligent camera 4 can, for example, recognize other lanes and associate other road users with lanes, as well as recognize traffic signs and land landmarks, which can be helpful in determining the precise location of the ego vehicle 2, for example, in relation to a high-resolution map. Furthermore, the ego vehicle preferably has a data storage device (not shown) configured to store the test run data 6 collected by the intelligent camera 4 and any other sensors that may be present. In Figure 1a, the test run data is represented by a file folder 6.

[0051] To encode the test run data 6, the so-called OSI documentation is used in particular. OSI stands for Open Simulation Interface, a general-purpose interface for automated operating functions to perceive the environment in a virtual scenario (https: / / opensimulationinterface.github.io / osi-documentation / ).

[0052] During (online) or after (post-) the test run, the test run data 6 is supplied to the system 10 for testing the driver assistance system, as indicated by the arrows in Figures 1a and 1b.

[0053] Figure 1b shows system 10 for testing the driver assistance system.

[0054] System 10 is preferably useful for evaluating the collected test run data 6 and for analyzing the driving behavior that the driver assistance system 1 should have exhibited during the test run in which the test run data 6 was generated.

[0055] The system 10 in Figure 1b is configured to implement method 100 for testing the driver assistance system 1 in Figure 2.

[0056] Therefore, the means 11 for assigning attribute Tx-yyy, the means 12 for inspecting the test run data 6, and the identification means 13 are preferably means configured to realize the respective assigned functions of the data processing system.

[0057] Means 14 for analyzing driving behavior may also be implemented within the data processing system. Preferably, in this case, it is also provided to simulate the driver assistance system 1, or to test only its software, particularly by a software-in-the-loop method.

[0058] However, more preferably, the means 14 for analyzing the driving behavior of the driver assistance system 1 is designed as a test stand, in particular as a vehicle test stand, a vehicle in-the-loop test stand, or a hardware in-the-loop test stand.

[0059] Preferably, the driver assistance system 1 is installed or connected to such a test stand 14, and the data fields of the test run data 6 corresponding to the identified scenario become available to the driver assistance system 1, or to sensors that supply information to the driver assistance system 1 via an appropriate interface. This is indicated by the arrow in Figure 1b.

[0060] In the case of the intelligent camera 4, such an interface may be one or more screens that display the environment around the vehicle to the camera 4 based on the data fields of the test drive data 6 corresponding to the scenario. In the case of radar, such an interface may be, for example, a radar target emulator. Alternatively, such an interface may also be provided to the test drive data 6, which is further processed, so that it can be directly provided only to the sensor chip of the driver assistance system 1 or the software of the sensor chip.

[0061] Preferably, the reactions or actions that characterize the driving behavior of the driver assistance system 1 are consequently supplied to the test stand 14 by further interfaces indicated by further arrows in Figure 1b.

[0062] The test stand 14 can analyze driving behavior based on parameters such as the driver assistance system 1 or control signals output by the driver assistance system 1 controlling the vehicle 2' on the test stand 14.

[0063] In detail, the driving behavior recorded by the driver assistance system 1 can be compared with reference data.

[0064] Instead of the exemplary embodiment of system 10 located outside the Ego Vehicle 2 to test the driver assistance system 1 shown in Figure 1b, system 10 may also be located inside the Ego Vehicle 2, for example, when the driver assistance system 1 is also located inside the Ego Vehicle 2 and the test driving data 6 is directly supplied by local sensors, particularly intelligent cameras 4.

[0065] Figure 2 is an exemplary embodiment of a computer-assisted method for testing the driver assistance system 1, which may be particularly implemented by the system 10 shown in Figure 1b.

[0066] In the first step of the procedure, the attribute Tx-yyy, recorded during the test run of Ego Vehicle 2 and therefore contained in the test run data, is assigned to the other vehicle. Thereafter, in the reference symbol of Tx-yyy, x represents the letters R, S, and A instead of "rear," "side," and "front." In each case, the symbol "y" represents the number indicating the lane for Ego Vehicle 2 and its position in the direction of travel.

[0067] An example assignment of attribute Tx-yyy to other road users is shown in Figure 3. Each row of the matrix shown in Figure 3 preferably corresponds to a lane, and therefore, the ego vehicle 2, depicted in black, is in the center lane.

[0068] Each road user around Ego Vehicle 2 is identified by an attribute beginning with T. The letters R, S, and A represent "behind," "behind," and "in front," respectively. The first digit after the hyphen indicates whether the other road user's lane is the same lane as Ego Vehicle's lane or a different lane. In the illustrated exemplary embodiment, the number 1 represents the lane to the right of Ego Vehicle 2, the number 2 represents the same lane as Ego Vehicle 2, and the number 3 represents the lane to the left of Ego Vehicle 2. In the illustrated exemplary embodiment, the last two digits after the hyphen represent the lane position of the road user in front of or behind the represented road user, which in this exemplary embodiment is a vehicle.

[0069] The attribute Tx-yyy is preferably assigned independently of the respective distances from ego vehicle 2 to other road users.

[0070] Each assigned attribute Tx-yyy reflects the relative position of another road user at the time of the test run data 6. Therefore, for each increment of time in which data is stored in the test run data 6, the attributes of other road users included are preferably also stored. Alternatively, to reduce data size, only one change to attribute Tx-yyy may be stored at a time.

[0071] Preferably, attribute Tx-yyy is assigned only within a defined distance from the ego vehicle 2. More preferably, this distance is within the measurement range of a sensor that detects the relative position of other road users to the ego vehicle 2. Preferably, as previously described, this sensor may be an intelligent camera 4.

[0072] The attribute Tx-yyy may further contain information about the direction of travel of another road user relative to Ego Vehicle 2. This may, for example, involve adding an additional character to the beginning of the attribute. For example, as shown in Figure 3, the character "o" (instead of "opposing") can identify an approaching vehicle by attribute oTA-101, and the character "c" (instead of "crossing") can identify a vehicle in crossing traffic by attribute cTA-302.

[0073] Figure 4 shows the time-sequenced progression of road users 3a and 3b with attribute Tx-yyy. Ego vehicle 2 is in the center lane.

[0074] The first road user 3a drives faster than ego vehicle 2 by changing lanes from the center lane to the right lane. Therefore, the attribute Tx-yyy of road user 3a changes from TA-201 to TA-301.

[0075] The second road user 3b is driving in the left lane behind ego vehicle 2 in the same lane as ego vehicle 2, and is also driving at a higher speed than ego vehicle 2, so is about to overtake ego vehicle 2. Correspondingly, when the second road user 3b is alongside ego vehicle 2 at a later point in time, the attribute of the second road user 3b changes from TR-101 to TS-101.

[0076] As already revealed, the distance da of the first road user 3a and the distance db of the second road user 3b preferably do not affect the assignment of attribute Tx-yyy. However, it is important that the second road user 3b moves from behind the ego vehicle 2 to alongside it, and the first road user 3a moves from the center lane to the right lane.

[0077] In the second step 102 of Method 100 shown in Figure 2, the test run data is examined to determine whether basic motions are occurring. This examination substantially constitutes a search of the test run data 6 to find known patterns of basic lateral motions LCL, LCR, IL and basic longitudinal motions GO, GC, FL. Furthermore, the search is preferably performed to find basic cornering motions. A requirement here is that for each basic motion, a pattern or model is defined that can be compared with the parameter profile and parameter set contained in the test run data 6. Such patterns can be stored, for example, as models. Preferably, these models can be generated using machine learning, in which case the models are preferably trained using test run data that has already been classified with respect to the basic motions. Therefore, preferably supervised machine learning is used, in which case a person classifies the test run data, and then an algorithm, such as an artificial neural network, is trained on this data.

[0078] The patterns generated in this manner are preferably stored in a list as predetermined basic motions and compared with the test run data 6 in step 102 of the inspection procedure.

[0079] Examples of basic lateral movements include "changing lanes to the left" (LCL), "changing lanes to the right" (LCR), "driving within the lane" (IL), "driving outside the lane," "changing lanes to the right," and "changing lanes to the left."

[0080] Examples of basic longitudinal movements include "initial start," "increasing the gap" (GO), "decreasing the gap" (GC), "following a vehicle," "driving in an empty lane" (FL), and "stopping."

[0081] Examples of basic cornering movements include "driving straight without curves," "cornering where the absolute value of curvature increases," "cornering at an exit where the absolute value of curvature decreases," "cornering with constant curvature," "left turn," "right turn," and "driving through a roundabout."

[0082] Referring to Figure 4, the second road user 3b is performing basic longitudinal motion FL in an empty lane and basic lateral motion IL while driving within the lane. In contrast, during the period of basic lateral motion shown, the first road user 3a first performs basic lateral motion IL while driving within the lane, then basic lateral motion LCL while changing lanes to the left, and finally returns to basic lateral motion IL while driving within the lane. The basic longitudinal motion shown by the first road user 3a throughout the entire period is driving FL in an empty lane.

[0083] In step 103 of the third procedure, the occurrence of predetermined scenarios during the test run is identified in the test run data. These scenarios preferably consist of a set of basic motions LCL, LCR, IL, GO, GC, FL and attribute Tx-yyy.

[0084] This allows for scenarios that only consider the basic motions of ego vehicle 2: LCL, LCR, IL, GO, GC, and FL. However, these scenarios typically involve ego vehicle 2 interacting with other road users 3a and 3b.

[0085] Examples of such predetermined scenarios include entering a lane in front of other road users 3a and 3b, entering a lane in front of the ego vehicle, overtaking other road users 3a and 3b, ego vehicle 2 being overtaken by other road users 3a and 3b, other road users 3a and 3b leaving the lane, and ego vehicle 2 leaving the lane.

[0086] The scenario may be freely defined, preferably by the test engineer, so that the attribute Tx-yyy and the basic motions LCL, LCR, IL, GO, GC, FL of the ego vehicle 2 and other road users 3a, 3b are combined into a predetermined scenario.

[0087] One example of such a scenario, namely the overtaking of Ego Vehicle 2, is shown in Figure 5, a structured diagram indicating the given attributes Tx-yyy and the basic motions LCL, LCR, IL, GO, GC, FL as a function of time t. More specifically, the diagram shows the changes in the basic longitudinal motions GO, GC, FL and basic lateral motions LCL, LCR, IL of the Ego Vehicle, and the changes in the basic longitudinal motions GO, GC, FL and basic lateral motions LCL, LCR, IL of the first road user 3a, represented as the vehicle in Figure 5b, as well as their respective given attributes Tx-yyy, as a function of time t.

[0088] Such groups or sequences of attribute Tx-yyy and basic motions LCL, LCR, IL, GO, GC, FL are determined during the analysis of the test run data 6, and data fields corresponding to the overtaking motion scenario of the ego vehicle 2 can be identified in the test run data 6.

[0089] The first road user 3a is performing basic lateral motion (IL) within the lane throughout the entire duration of the motion.

[0090] Ego vehicle 2 initially performs the basic lateral movement of staying in the lane (IL), but at 3 seconds, it begins to overtake, thereby initiating a left lane change (LCL). The lane change is completed at time t of 7 seconds. From this point until time t of 20 seconds, ego vehicle performs the basic lateral movement of staying in the lane (IL). At time t of 20 seconds, ego vehicle begins to change to the right lane by overtaking the first road user 3a, thereby initiating another basic lateral movement of changing to the right lane (LCR). This is completed at time t of 24 seconds. Ego vehicle 2 then returns to the right lane and continues the basic lateral movement of staying in the lane.

[0091] Correspondingly, the basic longitudinal motion of ego vehicle 2 progresses over time t. Initially, ego vehicle 2 approaches the first road user 3a, which is the basic longitudinal motion of a gap reduction GC. Since there are no other road users in the center lane, a state of longitudinal travel FL in the empty lane occurs as a result of the basic lateral motion of a lane change to the left LCL. Since there are no other road users ahead of the first road user 3a in the right lane, this state continues after a lane change to the right LCR.

[0092] On the other hand, the first road user 3a initially performs a basic longitudinal motion FL in the empty lane because the lane ahead is clear. Following the overtaking motion, the ego vehicle 2 enters the lane ahead of the first road user 3a and then pulls away from the first road user 3a at a higher speed in the same (right) lane, so the longitudinal motion changes to an expanding gap GO at a time t of 23 seconds.

[0093] The attributes assigned to the first road user 3a in relation to the ego vehicle are indicated in the bottom row of Figure 5a. The first road user 3a initially receives the attribute TA-101 because the first vehicle is ahead of the ego vehicle 2. After the ego vehicle makes a left lane change LCL, the first road user 3a is in the lane to the right of the ego vehicle's lane, so it receives the attribute TA-301. When the ego vehicle 2 catches up to and overtakes the first road user 3a, the first road user is next to the ego vehicle 2 in the lane to the right of the ego vehicle 2, so it receives the attribute TS-301. When the first road user 3a is completely overtaken by the ego vehicle 2, it is behind the ego vehicle 2 in the lane to the right of the ego vehicle 2, so it receives the attribute TR-301. If the ego vehicle changes lanes to the right and then returns to the right lane, the first road user 3a, being behind the ego vehicle 2 in the same lane, will receive the TR-101 attribute.

[0094] In the final step 104, the driving behavior of the driver assistance system in the identified scenarios is finally tested and analyzed based on the test drive data 6.

[0095] To this end, test runs are preferably conducted on the test stand 14 in identified scenarios, and the driver assistance system 1 and / or the vehicle 2' on which the driver assistance system 1 is installed are test-driven under conditions defined by test run data 6. The test run data 6 preferably includes road course, legal requirements from road signs, weather, topology, etc., in addition to boundary conditions derived from the positions of other road users 3a, 3b relative to the ego vehicle 2.

[0096] Preferably, during the test run, the operation or response of each driver assistance system being tested under given boundary conditions is observed or analyzed. This driving behavior of the driver assistance system is preferably compared with reference data to evaluate the driver assistance system 1 and, if necessary, to optimize calibration.

[0097] Preferably, the driver assistance system is test-driven based on data fields in the test drive data 6 that identify scenarios related exclusively to the driving behavior of the driver assistance system 1 being tested. This makes it possible to significantly reduce the time required to test the driver assistance system 1 or the distance of the test drive required.

[0098] As previously explained, this allows the test stand 14 to be designed not only as a vehicle test stand, but also as a test stand that simulates only essential parts of the vehicle 2' and / or the driver assistance system 1.

[0099] The exemplary embodiments described above are merely examples and are not intended to limit the scope of protection, uses, and configurations of the present invention. Rather, the foregoing description is intended to provide those skilled in the art with guidance for carrying out at least one exemplary embodiment, thereby enabling the creation of various modifications without departing from the scope of protection derived from equivalent combinations of claims and features, particularly with respect to the function and arrangement of the components described above. [Explanation of Symbols]

[0100] 1. Driver assistance system 2, 2' Ego Vehicle 3a, 3b Other road users 4 Camera Sensor / Camera 5 road 6. Test run data 10 Systems 11 Means for assigning attributes 12. Means for inspecting test run data 13 Means for identifying the occurrence of a predetermined scenario 14. Means / test benches for analyzing driving behavior LCL changing lanes to the left LCR (Lane Change to Right) IL (Lane driving) GO interval expansion Reduced GC interval FL Driving in an empty lane Tx-yyy attribute

Claims

1. A computer-based method (100) for testing the driver assistance system (1) of an ego vehicle (2) based on test driving data (6), In particular, step (101) assigns an attribute (Tx-yyy) to other vehicles (3a, 3b, ...) located in the immediate vicinity of the ego vehicle (2) and incorporated into the test run data (6), wherein the attribute (Tx-yyy) defines the relative position of each of the other vehicles (3a, 3b, ...) to the ego vehicle (2) at a time within the range of the test run data, and the attribute (Tx-yyy) is associated with the relevant time. Step (102) of examining the test drive data (6) to determine whether in each case basic lateral motion (LCL, LCR, IL) characterized by a vertical change in the position of one of the ego vehicle or the other vehicles relative to the road course, and in each case basic longitudinal motion (GO, GC, FL) characterized by a change in the forward and / or rearward distance of one of the ego vehicle (2) or the other vehicles (3a, 3b, ...), particularly in the same lane, is occurring, wherein the basic motion (LC Step (102) involves selecting L, LCR, IL, GO, GC, FL) from a predetermined list of basic motions (LCL, LCR, IL, GO, GC, FL), and associating the occurrence of the basic motion (LCL, LCR, IL, GO, GC, FL) with at least one relevant time point, and when inspecting the basic motion, a model is used that generates patterns based on test run data already classified with respect to the basic motion, specifically a model for recognizing the basic motions (LCL, LCR, IL, GO, GC, FL) in the test run data, Step (103) is to identify the occurrence of a predetermined scenario in the test run data (6) based on the basic motions (LCL, LCR, IL, GO, GC, FL) that have occurred, wherein the predetermined scenario is characterized by a set of basic motions (LCL, LCR, IL, GO, GC, FL) and attributes (Tx-yyy), In particular, step (104) of analyzing the driving behavior of the driver assistance system (1) exclusively in the identified scenario and Method (100), including the steps of the procedure.

2. The method according to claim 1 (100), wherein the test run data is retrieved for the attribute (Tx-yyy) and / or basic motion (LCL, LCR, IL, GO, GC, FL) contained in the predetermined scenario.

3. The method according to claim 1 or 2 (100), wherein a test run is performed on a test stand (14) using the test run data (6) in order to analyze the driving behavior of the driver assistance system (1) in the identified scenario.

4. The method according to claim 3 (100), wherein the test stand (14) is a vehicle test stand.

5. The method according to any one of claims 1 to 4 (100), wherein the list includes at least one of the basic lateral movement groups such as changing lanes to the left (LCL), changing lanes to the right (LCR), driving within a lane (IL), driving outside a lane, changing direction to the right, and changing direction to the left.

6. The method according to any one of claims 1 to 5 (100), wherein the list includes at least one of the basic longitudinal motion groups: initial start, increasing the distance (GO), decreasing the distance (GC), following a vehicle, driving in an empty lane (FL), and stopping.

7. The method according to any one of claims 1 to 6 (100), wherein the test drive data is further examined to determine whether a basic cornering motion has occurred, and the basic cornering motion is selected from a list that includes at least one of the basic cornering motion groups, such as straight driving without curvature, cornering with increasing absolute curvature, exit cornering with decreasing absolute curvature, cornering with constant curvature, left turn, right turn, and driving through a roundabout.

8. The method according to any one of claims 1 to 7 (100), wherein the attribute (Tx-yyy) indicates to the ego vehicle (2) whether another vehicle (3a, 3b, ...) is in the same lane, the right lane, or the left lane, and whether, with respect to the course of the road, the other vehicle (3a, 3b, ...) is in front of, behind, or alongside the ego vehicle (2).

9. The method (100) according to any one of claims 1 to 8, wherein the attribute (Tx-yyy) is assigned only to defined distances within the measurement range of a sensor (4) for determining the attribute of the ego vehicle (2), without depending on the distance (da, db, dc) of the other vehicles (3a, 3b, ...) to the ego vehicle (2).

10. The method according to any one of claims 1 to 9 (100), wherein the test run data is generated based on an actual test run, and the lanes of the ego vehicle (2) and the other vehicles (3a, 3b, ...) are determined by an intelligent camera (4).

11. The intelligent camera (4) is mounted on the ego vehicle (2), according to the method (100) of claim 10.

12. The method (100) according to claim 10 or 11, wherein known locations of land landmarks are further used for a reference system, such as a high-resolution map captured by the intelligent camera (4), in order to determine the lanes of the ego vehicle (2) and the other vehicles (3a, 3b, ...).

13. The method according to any one of claims 1 to 12 (100), wherein the test run data is generated based on actual test runs, and the relative positions of the other vehicles (3a, 3b, ...) with respect to the ego vehicle (2) are determined by an intelligent camera (4), a lidar and / or radar.

14. The method (100) according to claim 13, wherein the intelligent camera (4), lidar and / or radar are mounted on the ego vehicle (2).

15. A computer program product that, when executed by a computer, contains instructions prompting the computer to perform a step of the method described in any one of claims 1 to 14.

16. A computer-readable medium on which the computer program product described in claim 15 is stored.

17. A system (10) for testing a driver assistance system (1) based on test driving data (6) of an ego vehicle (2), In particular, means (11) for assigning attributes (Tx-yyy) to other vehicles (3a, 3b, ...) located in the immediate vicinity of the ego vehicle (2) and incorporated into the test run data, wherein the attributes (Tx-yyy) define the relative positions of each of the other vehicles (3a, 3b, ...) with respect to the ego vehicle (2) at a time within the range of the test run data, and the attributes (Tx-yyy) are associated with the relevant time, Means (12) for examining the test drive data (6) to determine whether, in each case, basic lateral motions (LCL, LCR, IL) characterized by a vertical change in the position of one of the ego vehicle (2) or the other vehicles (3a, 3b, ...) relative to the road course, and basic longitudinal motions (GO, GC, FL) characterized by a change in the forward and / or rearward distance of one of the ego vehicle (2) or the other vehicles (3a, 3b, ...) in each case, particularly in the same lane, are occurring, wherein the basic motions are selected from a predetermined list of basic motions, the occurrence of the basic motions (LCL, LCR, IL, GO, GC, FL) is also associated with at least one relevant point in time, and when examining the basic motions, a model for recognizing basic motions (LCL, LCR, IL, GO, GC, FL) in the test drive data is used, in particular, the means (12), wherein the basic motions are selected from a predetermined list of basic motions, the occurrence of the basic motions (LCL, LCR, IL, GO, GC, FL) is also associated with at least one relevant point in time, and when examining the basic motions, a model for recognizing patterns generated by machine learning based on test drive data already classified with respect to the basic motions, in particular, a model for recognizing basic motions (LCL, LCR, IL, GO, GC, FL) in the test drive data, Means (13) for identifying the occurrence of a predetermined scenario based on the occurrence of the basic motions (LCL, LCR, IL, GO, GC, FL), wherein the predetermined scenario is characterized by a set of basic motions and attributes, In particular, means (14) for analyzing the driving behavior of the driver assistance system (1) exclusively in the identified scenarios and A system (10) equipped with the following.

Images