• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    Method for validating a driver assistance system (3) of a vehicle, wherein tests (T) defined by test parameters (P) are run through for a given test scenario (4), during a first test (T (n)) at least one test parameter (P) is determined and for creation a second test (T (n + 1)) the first test (T (n)) is modified to shift the test parameter (P) within a critical area (7) associated therewith.
    公开号:AT513370A2
    申请号:T50729/2013
    申请日:2013-11-05
    公开日:2014-03-15
    发明作者:Peter Dipl Ing Priller
    申请人:Avl List Gmbh；
    IPC主号:
    专利说明:

    AV-3566 AT
    Virtual test optimization for driver assistance systems
    The present invention relates to a method for validating a driver assistance system of a vehicle, wherein tests are run through test parameters for a given test scenario.
    Driver assistance systems are now finding their way into almost all vehicle classes, with the stated goal of greatly reducing accidents or at least serious injuries and fatalities.
    Current driver assistance systems are each designed for specific driving situations, for example, an anti-lock braking system (ABS) counteracts possible blocking of the wheels, Electronic Stability Program (ESP) counteracts a possible break-out of the vehicle, Automatic Cruise Control (ACC) allows loose on-highway, assist Brake Assist the driver in emergency braking, Lane Keep Assist System (LKAS) helps the driver as a lane departure warning to keep track, for example, on highways, with many more driver assistance systems are known. Since only a driver assistance system is used in modern vehicles in the rarest of cases, newer vehicle generations refer to an extended assistance system, ie the combination of several driver assistance systems.
    It is important in the development or testing and validation of driver assistance systems to consider as many situations or test scenarios as possible, which require the intervention of a driver assistance system, in the development phase in order to be able to check the desired intervention of the driver assistance system already in the test phase.
    By way of example only, the dense inner-city traffic is cited, whose complexity makes it almost impossible to take into account all possible situations in the course of the test phase. Furthermore, it should be noted that the intervention of a driver assistance system may possibly lead to a new situation, which leads to the intervention of another driver assistance system. This can lead to new traffic scenarios that were not considered or considered in the original test phase.
    With regard to the formation of a wide variety of test scenarios for checking and optimizing driver assistance systems, DE 10 2011 088 807 A1, for example, shows a method in which a multiplicity of further test scenarios is created from a given test scenario by means of Monte Carlo simulation, ie a stochastic method. 2.18
    AV-3566 AT For each scenario created in this way, a course with, and a course without intervention of the driver assistance system is simulated. By comparing these two scenarios, it becomes possible to find quantitative measures of the effects of the intervention of the driver assistance system. For example, an accident risk, a risk of damage or the like can be quantified for each of the scenarios.
    The disadvantage is that a result of a simulated scenario has no effect on the simulation of another scenario and the scenarios are created exclusively by the stochastic method. The data resulting from a simulated scenario are therefore not used to improve, change or create another scenario.
    DE 10 2008 027 509 A1 shows a method for evaluating a driver assistance system with regard to its effectiveness already in the planning phase. For this purpose, a simulation based on the measurement data of a real accident is performed. At crucial points of the simulation, a sub-simulation is generated which involves the intervention of a driver assistance system. For example, this intervention may involve the activation of an automatic braking system with different delays. The results, or the outcome of the accident situation, are stored as a simulation data record. With regard to the base accident, whose data was used for the simulation, corresponding activation times for an automatic braking system can be calculated, for example, for different delays, which lead to an avoidance of the accident. In this way, a database of simulation data sets is created that can be used for a large number of driver assistance systems in order to obtain a reliable, real-data-based statement about the effectiveness of the driver assistance system.
    The disadvantage is that only measurement data of actually occurred accident situations is accessed. Scenarios for which no accident data are available, data from a driving situation in which there has been no loss of vehicle control, or from successfully prevented accidents or "almost" accidents, find no use in the presented method. Therefore, a set of measurement data, which might well be suitable for the creation of further test scenarios, discarded. Especially the critical area between an accident prevented by a driver assistance system and the occurrence of an accident or loss of control is that area which has the greatest potential for further development when testing driver assistance systems. The successful intervention of a driver assistance system is determined by its work in said critical area. 3.18
    AV-3566 AT
    It is therefore the object of the subject invention to enable extensive testing and validation of a driver assistance system.
    This object is achieved in that at least one test parameter is determined during a first test and the first test is modified to produce a second test in order to shift the test parameter within a critical range assigned to it. This ensures that especially critical areas within which a driver assistance system actively intervenes in the driving event are taken into account in the creation of further tests and tests whose outcome is initially rated as positive were also used for further investigation. By considering those critical areas and, based on the specific test parameters, selectively modifying the test, it is also possible to create tests that may possibly reveal hidden deficiencies in the driver assistance systems.
    An advantageous embodiment provides that the test parameter itself is changed so that it lies within a critical range. Thus, a test parameter, such as the speed, can be changed directly, resulting in a new starting situation for another test.
    A further advantageous embodiment provides that the test parameter is dependent on further test parameters and that these further test parameters are changed so that the test parameter is within a critical range. This makes it possible to indirectly influence a first test parameter and thereby to recognize possible relationships or the interaction of individual test parameters. Indirect influence can reveal interactions, which in turn lead to new possible tests.
    An advantageous embodiment provides that in order to produce the second test, the first test is modified such that the test parameter exceeds an assigned limit value. This deliberately provokes the activation of a driver assistance system, which subsequently leads to a new course of the second test. This makes it possible to increase the informative value of the second test, for example by deliberately or deliberately inducing a loss of control, since the effect or the effectiveness of the activated as a result of activated driver assistance system becomes apparent.
    A further advantageous embodiment provides that the tests in the real driving test and / or on corresponding test stands with an at least partially real vehicle and / or are carried out entirely virtual. This allows easy application 4/18
    AV-3566 AT of the process on a wide variety of existing, test or testing equipment.
    Advantageously, it is provided that the driver assistance system is formed from a plurality of individually acting driver assistance systems. As already mentioned in the introduction, this represents the usual state in today's vehicles. If this fact is also taken into account when testing and validating a driver assistance system, this also makes it possible to recognize and take into account the interaction of different systems or their mutual influence.
    A likewise advantageous embodiment provides that real driver or virtual sensor data are made available to the driver assistance system, that the driver assistance system calculates test parameters from the sensor data, then generates an internal driving situation and the internal driving situation is compared with the real or virtual driving situation. If sensor data was incorrectly processed / calculated by the driver assistance system, a certain difference results between the test parameters calculated for the internal driving situation and their actual values, which describe the driving situation. This allows early detection of non-recognized or incorrectly recognized or classified objects by the driver assistance system. As a result, possibly hidden errors in the driver assistance system, which affect the perception of the environment, can already be detected during the development phase. 20
    The subject invention will be explained in more detail below with reference to Figures 1 to 3, which show by way of example, schematically and not by way of limitation advantageous embodiments of the invention. It shows
    1 shows the relationship between the critical range, limit and position of a test parameter,
    2 shows the relationship between critical area, location of a test parameter, and a changing limit,
    3 shows the scheme according to the invention of the method for validating a driver assistance system. FIG. 4 shows a detail of the scheme shown in FIG. 3 with an advantageous one
    Complement. 5.18
    AV-3566 AT
    Subsequently, a number of terms will be used to describe the method according to the invention, which will be explained at the beginning.
    Test scenario 4 is understood to mean a certain frame condition / environment, for example passing through a curve by means of a vehicle. Other test scenarios 4 could include, for example, driving on inclines or slopes, or driving straight ahead with an obstacle or other road users on the roadway, and furthermore a variety of other test scenarios 4 are conceivable. All physical and dynamic indicators are referred to as test parameters P. The test parameters P include, for example, the roadway width, which radius of curvature has a curve, road characteristic values such as adhesion and friction values, roadway temperature, air humidity, wind strength and wind direction, with which speed the vehicle traverses the curve, which lateral acceleration the vehicle has, which slip on the wheels ( possibly at each individual), how large an obstacle is and where it is, where and how quickly other road users move, etc. Basically, it should be noted that the increase or decrease of a test parameter P, depending on its nature, different Can have an impact. For example, an increase in speed leads rather to a loss of control / accident, the increase in traction, however, increases the controllability of the vehicle.
    If further reference is made to the increase of a test parameter P, that change of the test parameter P is meant, which makes it possible to shift the test parameter P into its critical region 7, which will be described in more detail below.
    The test T (n) is formed from the combination of test scenario 4 and test parameter P. The test T (n) thus includes where (test scenario 4) a vehicle is to move under which conditions (test parameter P). A test T (n) could include driving on a low speed curve on a dry road. Another test T (n + 1) could involve driving on the same curve, with much higher speed, crosswind, and spot-on, ice-slippery road. A new test T (m) uses a new test scenario 4 and includes, for example, straight-ahead driving, downhill on wet roads.
    The test T (n) can be carried out either real, with a vehicle on a test track, a vehicle on a roller test bench with a virtual environment or completely virtual, in the form of a simulation, whereby hybrid forms are also conceivable. In a real test, test parameters P described above are determined by the driver assistance system 3 in a known manner via corresponding sensors. If the environment is virtually simulated, the sensor data are simulated accordingly. From these real or simulated sensor data 3 test parameters P are calculated by the driver assistance system. For example, the driver assistance system 3 by the real or simulated sensor data 6/18
    AV-3566 AT provided an instantaneous speed, calculated the driver assistance system 3 from the temporal change of a corresponding acceleration or deceleration. In this way also test parameters P are calculated by the driver assistance system 3. If an above-mentioned hybrid form is used, test parameters P are also calculated and / or made available.
    A driver assistance system 3 of a vehicle has, as already briefly mentioned in the introduction, the task of reducing accidents or at least serious injuries. A driver assistance system 3, without the intervention of the driver, actively intervenes in the driving event in order to avoid accidents and, above all, damage to persons as far as possible. As a driver assistance system 3, a combination of several driver assistants, so for example by anti-lock braking system (ABS) + Electronic Stability Program (ESP) + traction control (ASR) + Lane Keep Assist System (LKAS) can be seen. These possible combinations of different driver assistants are also referred to as driver assistance system 3.
    Some test parameters P, such as the lane width or curve radius of a curve, can be chosen freely and do not change during the test T (n). These are those test parameters P which are necessary in order to define the test scenario 4 more accurately for the test T (n), for example the specification of the curve radius, if the test scenario 4 is selected as the passage through a curve. Test parameters P such as, for example, the speed of the vehicle are initially freely specified, but may change during the test T (n) since, for example, a driver assistance system 3 reduces the speed.
    It is therefore understandable that test parameters P need not necessarily be predetermined. They can also arise as a result of very different relationships, or change during the test T (n). For example, if the vehicle is spinning, the lateral acceleration of the vehicle changes during the test T (n), although this was initially undetermined or not predetermined. Some test parameters P can not be selected "directly", but only indirectly influenced, such as, for example, the slip, which depends inter alia on the driving torque of the vehicle and the coefficient of friction between the wheels and the road surface. Also by the intervention of a driver assistance system 3, test parameters P, as already mentioned using the example of the speed, can be changed continuously during the test.
    In this context, the driving situation 5 is spoken. The driving situation 5 includes that state, the position of the vehicle, which sets during the test T (n). It results from the given, or during the test T (n) 7/18
    AV-3566 AT resulting test parameter P. This driving situation 5, as already stated, for example, by a higher lateral acceleration than it was previously defined as test parameter P be marked. Such a driving situation 5 could stop in the test T (n), which involves driving on a high-speed, crosswind and icy 5-lane curve, since the vehicle is skidding.
    The critical area 7 of a test parameter P is that area between controlled driving behavior and an undesired driving situation 5. Controlled driving behavior is, for example, a driving behavior in which the intervention of a driver assistance system 3 is not necessary. As unwanted driving situation 5, for example, the touch of another 10 vehicle, a curb, etc., or the partial or total loss of control is called. The critical area 7 is limited by a lower limit 20 and an upper limit 40 for the test parameter P. Below the lower limit 20 of the critical area 7, the driving behavior is controlled and activation of a driver assistance system 3 is not necessary. Although a driver assistance system 3 is already activated above the upper limit 40, the avoidance of an undesired driving situation 5, an accident, or the recurrence of the control is no longer possible.
    Between lower limit 20 and upper limit 40 is said critical area 7 within which there may be loss of control, a driver assistance system 3 can be activated 20 and by the assistance of the driver assistance system 3 control can be restored.
    Thus, test parameters P are assigned a critical area 7 in the form of a lower limit 20 and an upper limit 40, within which a driving situation 5 usually occurs, which may require the intervention of a driver assistance system 3, for example a certain amount of forces acting on it one of the tires and which would lead to loss of traction. These critical areas 7 or their lower limits 20 and upper limit 40, for example, by fixed values, characteristics, maps, and the like, given for the individual test parameters P, or even freely chosen. However, the critical area 7 of a test parameter P is not necessarily coupled with the activation of the driver assistance system 3. A test parameter P can thus be within its critical range, yet the driver assistance system 3 has not yet been activated. The activation of the driver assistance system 3 is coupled to the limit value G of a test parameter P that is within the critical range 7. 18.8
    AV-3566 AT
    The limit value G of a test parameter P is the value at whose reaching or exceeding the driver assistance system 3 is activated for assistance. The limit value G is set by the driver assistance system 3 using diagrams, maps, calculation formulas and the like which are based, for example, on already known driving situations, but may also vary during a test T (n), as described below. This limit value G is within the critical range 7. The position of the limit value G is dependent on the selected and / or during a test T (n) adjusting test parameters P and therefore may also change in the course of a test T (n). For example, an Automatic Cruise Control (ACC) ensures that the traffic in the column is kept at a constant distance from the vehicle in front. The limit G, at which an automatic braking system is activated, is initially dependent on the distance and the speed with which both vehicles move. If increased slippage is suddenly detected on one of the wheels of the vehicle, the driver assistance system 3 starts from changed road conditions and the limit value G at which an automatic brake system is activated is correspondingly reduced in order to avoid a rear-end collision even on a "slippery" road.
    As a result of the interaction of the individual test parameters P with one another, not only the test parameters P themselves, but also their limit values G can be influenced. Although these relationships can be represented in maps, it can nevertheless be assumed that not all relationships can be completely recorded for the driver assistance system 3.
    The limit value G does not necessarily lie at the lower limit 20 of the critical area 7, so that a driver assistance system is activated as soon as a test parameter P comes within the critical range. For example, it is conceivable that the limit value G for the test parameter P "slip lies approximately in the center of its associated critical region 7. The slip can therefore assume values during the test T (n) which, although they are within its critical range, through which its limit value G has not yet been exceeded, and therefore a driver assistance system 3, such as an anti-slip control (ASR), does not yet Help is activated. By way of the position of the limit value G within the critical region 7, a certain tolerance 30 is defined, via which it is selected to what extent a test parameter P is critical, but assistance by a driver assistance system 3 is "not necessary".
    FIG. 1 shows the relationship between critical region 7, its lower limit 20 and upper limit 40, tolerance 30, limit value G and position of a test parameter P. A "snapshot" is shown, since, as already explained, the position of a test parameter P and the position of the associated limit G in the course of the test T (n) constantly 9/18
    AV-3566 AT can change. By way of example, the position of the limit value G, initially selected by the driver assistance system 3 using diagrams, performance maps, calculation formulas and the like which are based, for example, on already known driving situations, is selected in the center of the critical region 7, whereby any other position of the limit value G is also determined by the Driver assistance system 3 can be selected. The closer the limit value G is to the lower limit 20 of the critical region 7, the smaller the tolerance 30. Within the tolerance 30, a test parameter P is indeed critical, but a driver assistance system 3 is "not" activated and assistance remains. By way of example, the test parameter P is exactly within this range. If the slip on one of the wheels of the vehicle is taken as the test parameter P in this case, the position of the test parameter P within the tolerance 30 means that there is already some slippage, but this is so small that assistance by the driver assistance system 3 is not necessary yet.
    FIG. 2 again shows the relationship in the form of a characteristic diagram, between critical region 7, its lower limit 20 and upper limit 40, and the limit value G for two test parameters P, which are dependent on one another, with the possibility of shifting the limit value G within one test T (n) is shown at two different times t1 and t2. For example, the test parameter P (v) represents the speed of a first vehicle and the test parameter P (a) the distance to a second vehicle ahead of it. The relationship between the two test parameters P (v) and P (a) is represented by the point X marked in the map. The higher the speed of the first vehicle, ie the test parameter P (v), the greater must be the distance, ie the test parameter P (a), to the preceding vehicle, so that point X does not come into the critical area 7 or Limit G exceeds. Although the point X lies in the critical region 7 at a first time t1, the limit value G (t1) has not yet been exceeded, so a driver assistance system 3, for example an automatic brake system, has not yet been activated. At another time t2, the speed of the first vehicle, that is to say the test parameter P (v), and the distance to the preceding vehicle, that is to say the test parameter P (a), are not changed and the position of the point X remains the same. The fact that the road conditions at the point where the first vehicle is at the time t2 is different from the road conditions at the place where the first vehicle was at the time t1, for example, due to deterioration of the road conditions due to moisture, is different but also the position of the limit G (t2) from the position of the original limit G (t1). Point X has therefore exceeded the limit G (t2) and the driver assistance system 3 has been activated. This will again ver-10/18
    AV-3566 AT clarifies that the interaction of the individual test parameters P can influence not only the test parameters P themselves, but also limit values G.
    FIG. 3 shows a schematic of the method according to the invention for testing and validating a driver assistance system 3 of a vehicle. In a database 2, for example, various conditions / environments (test scenarios 4) are stored. From the database 2, an environment is used as test scenario 4. Based on the test scenario 4, a test T (n) is carried out in which either initially no intervention of a driver assistance system 3 is necessary, or the driver assistance system 3 to be tested, in accordance with its task, acts as a supportive vehicle. The test T (n) is defined by test parameters P already described, such as speed, traction, distance to the roadway, etc., in combination with the test scenario 4.
    For this purpose, in a real test, the driving situation 5 in which the vehicle is located, that is to say the predetermined test parameter P resulting from the test T (n), is determined by the driver assistance system 3 via corresponding sensors as already described. If the environment is virtually simulated, the driver assistance system 3 is provided with correspondingly simulated sensor data or the test parameters P. The driving situation 5 can therefore include real sensor data and / or virtually created data. Based on the driving situation 5, should this be necessary in the context of the test T (n), the driver assistance system 3 takes appropriate measures to assist the vehicle. As a result of these measures, the test T (n) is influenced accordingly in its course, or test parameter P is changed.
    If the test T (n) is carried out for the test scenario 4, a certain driving situation 5 results which, as already explained, is defined by the test parameter P selected or resulting during the test T (n). During the test T (n), the test parameters P are determined, in a real test T (n), for example, read from a control unit or measured by sensors. Based on the critical regions 7 associated with the test parameters P, an evaluation 6 of the test parameters P is carried out.
    As evaluation 6, the comparison of the test parameters P, which resulted from the driving situation 5 during the test T (n), with the critical areas 7 assigned to them is understood. The evaluation 6 may, for example, show that a test parameter P has remained below its critical range 7 during the test T (n), thus has not reached its limit value G, and therefore the driver assistance system 3 has not been used. The result of the evaluation 6 thus contains which "locations" the test parameters P are within or outside the critical area 7 with respect to the lower 11/18 thereof
    AV-3566 AT
    Limit 20, upper limit 40 and the limit G during the test T (n) have taken. If, in the course of the test T (n), the exceeding of the upper limit 40 of a critical area 7 is detected, and if control loss, intrusion into the oncoming traffic area, collision with another vehicle or another undesired driving situation 5 has occurred the test T (n) for the test scenario 4 as "failed".
    In principle, a driver assistance system 3 should make it possible to avoid a loss of control, an accident, a collision with other vehicles or objects, etc.
    However, if the test T (n) has not been passed as described above, then an error of the driver assistance system 3 could possibly exist, or at least potential exist for improving and / or further developing the driver assistance system 3. In order to be able to analyze the reasons for the "failure" of the driver assistance system 3, all the test parameters P which have been selected and / or which have been produced during the test T (n) will be able to better recognize relationships , also the test scenario 4 subjected to an analysis 8, whereby possible errors and / or weak points of the driver assistance system 3 can be detected.
    If the test T (n) is terminated in the form that, for example, none of the test parameters P has come within its critical range 7 or at least one of the test parameters has come within its critical range 7 but has not exceeded its limit value G, all test parameters will become P is stored in a result database 10 after its evaluation 6. Even if a test parameter P has exceeded its limit value G, but the driver assistance system 3 has intervened supportively and successfully and thus corresponding directional stability, timely braking to standstill, etc. has been realized, all test parameters P are stored in a result database 10 after their evaluation 6. There, the test parameters P are still available for possible later analyzes, for documentation, as a basis for further developments or the like.
    Furthermore, in these cases, the test parameters P are used for their modification 6 for a modification 9. Using this, the modified test T (n + 1) is formed, which works with the original, same test scenario 4, which was also used for test T (n), with the difference that an arbitrary test parameter P, which occurs in the course of the test Test T (n) has been located below the lower limit 20 or within its critical range 7, where the test T (n + 1) is likely to be within its critical range 7 or even exceeds its limit value G.
    For this purpose, as already mentioned, some test parameters P can be changed directly. However, as already mentioned, there is also the possibility that a first test parameter-12/18
    AV-3566 AT ter P of further test parameters P is dependent. In order to ensure with high probability that the first test parameter P is within its critical range 7, these test parameters P, which influence the first test parameter P, can accordingly also be changed, if appropriate directly. The process of modification 9 takes place in that, for example, said first test parameter P, if this is possible directly, or those test parameters P on which it is dependent, is iteratively changed step by step. In this context, iterative means that a modification 9 takes place between the individual tests T (n), T (n + 1), T (n + 2),... And thus test parameters P are shifted step by step. However, the same test parameter P does not always have to be changed by the modification 9.
    Because the process is carried out iteratively, the modification 9 is preferably applied to those test parameters P which were within their critical range 7 during the test T (n) but have not yet exceeded their limit value G. Since a said test parameter P yes is already in the critical range 7, possibly only a slight modification 9 is sufficient to exceed its limit value G. It should be noted, however, that it is not possible to predict a safe overshoot, or even exact achievement of the limit value G, because, as already stated, due to a variety of relationships between the test parameters P, their limit values G can also be shifted, or these are not yet known ,
    The thus formed new test T (n + 1) therefore preferably corresponds to a test T (n) in which a test parameter P which has not exceeded its limit value G during test T (n) now exceeds its limit value G, ie in the range between controlled driving behavior and loss of control, with a driver assistance system 3 actively assisting. This is exactly the area that carries the greatest potential for further development when testing driver assistance systems 3. It can be seen whether the driver assistance system 3 can intervene as far as helping, so that even with test T (n + 1), in which the driver assistance system 3 is tested with high probability in the critical range, no loss of control or accident or any unwanted Driving situation 5 comes and the test T (n + 1) is thus passed. With "high probability" because the displacement of the test parameter (s) can be iterated, as already described, and it can not necessarily be ascertained whether a test parameter P actually lies within its critical range 7 or even reaches or exceeds its limit value G. , If the corresponding test parameter has not reached or exceeded its limit value G, a new test T (n + 1) has been performed in which other test parameters may have reached or exceeded their limit. 13/18
    AV-3566 AT
    Should it not be possible to pass the test T (n + 1) based on a test scenario 4 despite the intervention of the driver assistance system 3, an analysis 8 is carried out, as already described, since a weak point of the driver assistance system 3 may have been found. The analysis 8 can subsequently trigger a correction of errors already occurring in the development phase of the driver assistance system 3.
    Of course, the described modification 9 of the test T (n) can be arbitrary often / long, for which reason corresponding termination criteria, such as a maximum number of modifications 9 of a test T (n) or a maximum test time, can be provided. If said abort criterion is met, another environment stored in the database 2 can be used as a new test scenario 4 and a new test T (m) can be carried out. Also for test T (m) modifications of test parameters P are carried out again and thereby further tests T (m + 1), T (m + 2), ... are formed.
    Figure 4 shows a section of the scheme shown in Figure 3 with an advantageous addition. As already described, in a real test, the driving situation 5 in which the vehicle is located is determined by the driver assistance system 3 in a known manner via corresponding sensors. If the environment is virtually simulated, the driver assistance system 3 is provided with correspondingly simulated sensor data. From these real or simulated sensor data, the driver assistance system 3 calculates test parameters P or creates an internal driving situation 31 based on the sum of the data provided to the driver assistance system 3.
    If a driver assistance system 3 could not prevent an undesired driving situation 5, such as a collision, or if the driver assistance system 3 did not intervene despite the need, or was not activated, a possible source of error is that the driving situation 5 is incorrectly selected by the driver assistance system 3 " was estimated ". This would indicate that the internal driving situation 31 does not correspond to the actual or simulated driving situation 5. The reason could be, for example, that the sensor data from the driver assistance system 3 are processed incorrectly. This results in a certain difference between the test parameters P calculated for the internal driving situation 31 and their actual values which describe the driving situation 5.
    In order to detect such possible errors, as shown in FIG. 3, the internal driving situation 31 and the real or simulated driving situation 5 can be supplied to a comparison 11. If comparison 11 results in a difference between internal driving situation 31 and actual or simulated driving situation 5, and this difference exceeds a permissible, freely selectable value, the result of the comparison becomes 14/18
    AV-3566 AT 11 an analysis 8 subjected. As a result, in turn, possible errors and / or weak points of the driver assistance system 3 can be detected and their rectification can already take place in the development phase of the driver assistance system 3. 15/18
    权利要求:
    Claims (7)
    [1]
    A method for validating a driver assistance system (3) of a vehicle, wherein for a given test scenario (4) tests (T) defined by test parameters (P) are passed through, characterized in that during a first test (T (n)) at least one test parameter (P) is determined and to establish a second test (T (n + 1)), the first test (T (n)) is modified to match the test parameter (P) within an associated one critical area (7).
    [2]
    2. The method according to claim 1, characterized in that the test parameter (P) 10 itself is changed so that it lies within a critical range (7).
    [3]
    3. The method according to claim 1, characterized in that the test parameter (P) of further test parameters (P) is dependent and that these further test parameters (P) are changed so that the test parameter (P) is within a critical range (7) ,
    [4]
    4. The method according to any one of claims 1 to 3, characterized in that for the preparation of the second test (T (n + 1)), the first test (T (n)) is modified so that the test parameter (P) has a limit (G) exceeds.
    [5]
    5. The method according to any one of claims 1 to 4, characterized in that the tests (T) in the real driving test and / or on corresponding test benches with an at least partially real vehicle and / or are carried out entirely virtual.
    [6]
    6. The method according to any one of claims 1 to 5, characterized in that the driver assistance system (3) consists of several, individually acting driver assistance systems is formed.
    [7]
    7. Method according to one of claims 1 to 6, characterized in that the driver assistance system (3) real or virtual sensor data are made available, 25 that the driver assistance system (3) from the sensor data test parameter (P) calculated, from an internal driving situation ( 31) and the internal driving situation (31) is compared with the real or virtual driving situation (5). 16/18
    类似技术:
    公开号 | 公开日 | 专利标题
    AT513370B1|2015-11-15|Virtual test optimization for driver assistance systems
    DE19823303A1|1998-12-10|Adaptive control for road vehicle motion
    DE102010008258A1|2011-08-18|Method for the automatic prevention of aquaplaning
    WO2000032456A1|2000-06-08|Method and device for determining the adhesion and adhesion limit for vehicle tyres
    DE102009006299A1|2010-02-11|Vehicle travel control system for autonomous braking
    DE102007029870A1|2009-01-02|Tire condition monitoring method and apparatus
    DE10320809A1|2004-11-25|Car motion recognition and monitoring procedure processes data from acceleration, speed, force and body noise sensors using pattern recognition based on state vectors
    DE102015103570A1|2015-09-24|Driver Anomaly Detection
    DE102007039039A1|2009-01-22|Control of safety means of a motor vehicle
    DE102015116469A1|2016-04-07|DEHUMIDIFICATION OF VEHICLE FORWARD AFTER IMPACT
    DE102011088805A1|2013-06-20|Method for developing and/or testing of driver assistance system for motor vehicle, involves determining several scenarios in modeling of prior collision phase by using Monte Carlo simulation based on driving situation
    DE102014226211A1|2016-06-23|Method for controlling and / or regulating a brake system for a vehicle and brake system
    DE102008051962A1|2009-05-14|A reconfigurable structure method for estimating the lateral velocity of a vehicle
    DE102016211728A1|2018-01-04|Reibwertschätzer
    DE102013008073A1|2014-11-13|Method and device for collision avoidance or collision sequence reduction in trailer operation
    EP1966570A1|2008-09-10|Method for checking the plausibility of a determined vehicle mass
    DE102008027526B4|2015-10-29|Method for the prognostic evaluation of a predictive safety system of a motor vehicle
    WO2003039929A1|2003-05-15|Method and device for treating suspected errors
    DE102013114172A1|2014-12-11|Apparatus and method for determining a Kurzzeitfahrtendenz
    EP1294599B1|2011-02-02|Method for improving the controlling action of an abs
    DE112018005154T5|2020-07-02|METHOD FOR OPTIMIZING A VEHICLE SPEED INDICATOR, WHICH IS PROVIDED FOR THE STEERING ASSISTANCE FUNCTIONS AND THE SAFETY FUNCTIONS
    DE102009026813A1|2010-12-09|Method for generating a differential torque acting on the vehicle wheels of a vehicle
    DE102013011623A1|2015-01-15|Method and device for the automatic control of a longitudinal dynamics of a motor vehicle
    DE102013011624A1|2015-01-15|Method and device for the automatic control of a longitudinal dynamics of a motor vehicle
    DE102010038463A1|2012-02-02|Method for recognizing dynamic mechanical loads of given rating classes during driving of vehicle, involves recognizing dynamic loading of rating class for waveforms, if determined similarity meets predetermined conditions
    同族专利:
    公开号 | 公开日
    JP6559127B2|2019-08-14|
    AT513370A3|2014-07-15|
    US9937930B2|2018-04-10|
    EP3066529A1|2016-09-14|
    US20160280233A1|2016-09-29|
    CN105849657B|2019-01-18|
    KR20160079839A|2016-07-06|
    CN105849657A|2016-08-10|
    AT513370B1|2015-11-15|
    KR102234064B1|2021-03-31|
    WO2015067649A1|2015-05-14|
    JP2016537626A|2016-12-01|
    EP3066529B1|2020-01-01|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    JPH05256734A|1991-09-09|1993-10-05|Banzai:Kk|Composite testing apparatus for running function of vehicle|
    JPH08122218A|1994-10-19|1996-05-17|Nissan Altia Co Ltd|Multifunctional testing device for vehicle performance|
    JP3677104B2|1995-10-31|2005-07-27|安全自動車株式会社|Composite test apparatus and electric inertia control method thereof|
    DE19545001B4|1995-12-02|2005-03-24|Wabco Gmbh & Co.Ohg|Method for yaw moment attenuation in an anti-lock braking system|
    JP2773107B2|1996-06-20|1998-07-09|豊田通商株式会社|Driving simulator device|
    CN1159187C|1998-07-01|2004-07-28|雅马哈发动机株式会社|Auxiliary power driven vehicle|
    DE19934376A1|1999-07-22|2001-01-25|Wabco Gmbh & Co Ohg|Traction control method|
    JP3811366B2|2000-04-21|2006-08-16|住友ゴム工業株式会社|Road surface information distribution system, vehicle information aggregation and distribution system, vehicle information transmission device, and vehicle control program|
    JP4211330B2|2002-09-04|2009-01-21|日産自動車株式会社|Development support apparatus and development support method for anti-lock brake system for vehicle|
    DE10355022B4|2003-11-25|2016-10-06|Prof. Dr.Thomas Bartsch|Method for monitoring a technical system|
    DE102005036953A1|2005-08-05|2007-02-08|Robert Bosch Gmbh|Method for generating environmental hypotheses for driver assistance functions|
    JP2007245901A|2006-03-15|2007-09-27|Hitachi Ltd|Vehicular motion control apparatus|
    DE502006008342D1|2006-09-11|2010-12-30|Dspace Gmbh|Method for testing an electronic control system|
    DE102006044086B4|2006-09-20|2013-05-29|Audi Ag|System and method for the simulation of traffic situations, in particular accident-critical dangerous situations, as well as a driving simulator|
    DE102007053500A1|2007-06-11|2008-12-18|Audi Ag|Safety system assessing method for motor vehicle, involves assessing safety system based on reconstruction of accidents with respect to system, and performing realistic road trial with real and virtual components during reconstruction|
    DE102007039617B4|2007-08-22|2013-05-29|Continental Automotive Gmbh|Automatic mirror adjustment in a vehicle|
    US8706376B2|2008-05-23|2014-04-22|Bosch Corporation|Vehicular ABS control system with internal parameter automatic calibration function|
    US20090300422A1|2008-05-30|2009-12-03|Caterpillar Inc.|Analysis method and system using virtual sensors|
    DE102008027509A1|2008-06-10|2009-12-31|Audi Ag|Method for the prognostic evaluation of at least one predictive safety system of a motor vehicle|
    AT505326B1|2008-09-04|2010-01-15|Avl List Gmbh|DEVICE FOR THE MODEL-BASED DEVELOPMENT OF A COMBUSTION ENGINE|
    WO2010038317A1|2008-09-30|2010-04-08|Nissan Motor Co., Ltd.|System provided with an assistance-controller for assisting an operator of the system, control-operation assisting device, control-operation assisting method, driving-operation assisting device, and driving-operation assisting method|
    CN101482451A|2008-12-17|2009-07-15|奇瑞汽车股份有限公司|Entire car road virtual pilot system and its test method|
    AT10759U3|2009-04-23|2010-07-15|Avl List Gmbh|METHOD AND DEVICE FOR VERIFYING AN AUTOMATION SYSTEM|
    CN102426425B|2011-11-08|2013-07-17|重庆邮电大学|Automobile ABS virtual reality simulation system|
    DE102011121260A1|2011-12-15|2013-06-20|Gm Global Technology Operations, Llc|A method for assisting a driver of a motor vehicle in the event of loss of attention with the aid of an error counter|
    DE102011088805A1|2011-12-16|2013-06-20|Bayerische Motoren Werke Aktiengesellschaft|Method for developing and/or testing of driver assistance system for motor vehicle, involves determining several scenarios in modeling of prior collision phase by using Monte Carlo simulation based on driving situation|
    DE102011088807A1|2011-12-16|2013-06-20|Bayerische Motoren Werke Aktiengesellschaft|Method for developing and/or testing driver assistance system for motor car, involves determining set of scenarios and characterizing set of scenarios by parameter that is related to driving conditions of car|US10599818B2|2012-10-02|2020-03-24|Banjo, Inc.|Event-based vehicle operation and event remediation|
    EP3309721A1|2016-09-23|2018-04-18|KPIT Technologies Ltd.|Autonomous system validation|
    CN107479532A|2017-08-04|2017-12-15|安徽江淮汽车集团股份有限公司|The domain controller test system and method for a kind of intelligent automobile|
    US10345811B2|2017-11-14|2019-07-09|GM Global Technology Operations LLC|Method and apparatus for scenario generation and parametric sweeps for the development and evaluation of autonomous driving systems|
    AT521713A2|2018-10-11|2020-04-15|Avl List Gmbh|Procedure for the detection of security-relevant data flows|
    AT521607B1|2018-10-24|2020-03-15|Avl List Gmbh|Method and device for testing a driver assistance system|
    DE102018128890A1|2018-11-16|2020-05-20|Bayerische Motoren Werke Aktiengesellschaft|Method and test device for testing a driver assistance system for a motor vehicle|
    CN109991024B|2019-04-23|2020-12-29|重庆长安汽车股份有限公司|Three-level automatic driving vehicle over-bending capability test method|
    CN111537236B|2020-04-24|2021-04-30|吉林大学|Traffic jam auxiliary system testing method|
    DE102020120141A1|2020-07-30|2022-02-03|Iav Gmbh Ingenieurgesellschaft Auto Und Verkehr|Procedure for optimizing tests of control systems for automated vehicle dynamics systems using probabilistically predicted system responses|
    CN112711260B|2020-12-29|2022-02-11|清华大学苏州汽车研究院（相城）|Expected function safety test evaluation method for error/omission recognition of automatic driving vehicle|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    ATA50729/2013A|AT513370B1|2013-11-05|2013-11-05|Virtual test optimization for driver assistance systems|ATA50729/2013A| AT513370B1|2013-11-05|2013-11-05|Virtual test optimization for driver assistance systems|
    EP14796034.8A| EP3066529B1|2013-11-05|2014-11-05|Virtual test optimization for driver assistance systems|
    KR1020167014111A| KR102234064B1|2013-11-05|2014-11-05|Virtual Test Optimization for Driver Assistance Systems|
    PCT/EP2014/073797| WO2015067649A1|2013-11-05|2014-11-05|Virtual test optimization for driver assistance systems|
    US15/034,645| US9937930B2|2013-11-05|2014-11-05|Virtual test optimization for driver assistance systems|
    JP2016528140A| JP6559127B2|2013-11-05|2014-11-05|Virtual test optimization of driver assistance system|
    CN201480071221.6A| CN105849657B|2013-11-05|2014-11-05|Virtual test for driving assistance system optimizes|
    [返回顶部]