Method and device for assessing the criticality of a traffic situation

Abstract

A first general aspect of the present disclosure relates to a method for assessing the criticality of a traffic situation. In the traffic situation, at least part of the surroundings of a first road user is obscured, and a second road user may be located in the obscured part of the surroundings. The first road user is a vehicle. The method includes accessing a motion model for a traffic situation with a first road user and a second road user. The second road user is obscured from the first road user. The motion model models a future movement of the first and second road users. The method further includes processing the motion model to assess the criticality of the traffic situation and outputting the criticality of the traffic situation.

Description

background

[0001] The assessment of critical traffic situations is necessary and well-established in numerous technical contexts. For example, when testing automated vehicles, it can be evaluated whether critical traffic situations arise for a vehicle under development or for a vehicle component. During vehicle operation, the assessment of critical traffic situations can also be used both to warn vehicle occupants and to control the vehicle.

[0002] Various metrics have been proposed for criticality assessment. For example, some metrics can be used to evaluate criticality in different traffic situations involving two road users. One example is the "SafeLatLon" metric ("SLL metric" - described in German patent application DE 102023211082.2). The SLL metric uses motion models for the two road users and considers a so-called "reasonable worst-case scenario" (e.g., assuming that a road user ahead brakes with maximum force). One advantage of such model-based metrics is that the criticality assessments are understandable and explainable.

[0003] Existing techniques for assessing critical traffic situations may be limited to specific traffic scenarios. However, vehicles can encounter a wide variety of traffic situations in the field. It is therefore desirable to provide techniques for assessing this diverse range of traffic situations. Summary

[0004] A first general aspect of the present disclosure relates to a method for assessing the criticality of a traffic situation. In the traffic situation, at least part of the surroundings of a first road user is obscured, and a second road user may be located in the obscured part of the surroundings. The first road user is a vehicle. The method includes accessing a motion model for a traffic situation with a first road user and a second road user. The second road user is obscured from the first road user. The motion model models a future movement of the first and second road users. The method further includes processing the motion model to assess the criticality of the traffic situation and outputting the criticality of the traffic situation.

[0005] A second general aspect of the present disclosure relates to a method for generating control information for a vehicle in a traffic situation. In the traffic situation, at least part of the environment is obscured for a first road user, and a second road user may be located in the obscured part of the environment. The first road user is the vehicle. The method includes accessing a motion model for a traffic situation with a first road user and a second road user, the second road user being obscured for the first road user. The motion model models a future movement of the first and second road users. The method further includes processing the motion model to generate control information for the vehicle (e.g.,so that the criticality of the traffic situation is below a certain threshold) and output of the control information for the vehicle.

[0006] A third general aspect of the present disclosure relates to an environment designed to perform one of the methods described in the first and second aspects. The environment may be a test and / or development environment for vehicles and / or vehicle components. In other examples, the environment may be or contain a vehicle or a vehicle component.

[0007] A fourth general aspect of the present disclosure relates to a computer program containing instructions which, when executed by a computing unit, cause the computing unit to perform a procedure according to the first aspect or the second aspect.

[0008] A fifth general aspect of the present disclosure relates to a computer-readable medium or signal that stores and / or contains the computer program according to the fourth aspect.

[0009] The techniques of the first, second, third, fourth and fifth general aspects may have one or more of the following advantages in certain situations.

[0010] First, the criticality of a traffic situation can be assessed even if potentially obscured road users are present. An example of this is driving on a highway where a truck is in the field of vision of a first road user, thus obscuring part of the road. The second road user can then drive in the truck's shadow. If the first road user then overtakes the truck, this can lead to critical situations. Some known metrics can only assess situations with two dynamic (e.g., moving) road users if there is a line of sight between them (e.g., the SLL metric mentioned above). The techniques of this disclosure can make it possible to assess the criticality of situations with obscured road users as well. The assessments generated in this way can be used in a variety of ways. For one, an occupant (e.g.,A driver of the vehicle can be warned in a situation deemed critical (and subsequently intervene in the vehicle's driving behavior accordingly). In other examples, the ratings can be used for selecting or marking test or training data. Additionally or alternatively, the ratings can be used during testing or training itself. In this way, the effort required to test (e.g., validate) a vehicle or one of its components can, in some cases, be (significantly) reduced.

[0011] Secondly, the motion models or criticality metrics of this disclosure can be used to generate control information for a vehicle. In these cases, the criticality of a traffic situation is not assessed (or not only assessed) based on the motion models or criticality metrics, but rather control information is calculated to ensure, for example, that a criticality threshold is not exceeded during operation. This control information can then be used in some examples to control the vehicle more safely.

[0012] Some terms are used in the present revelation in the following way.

[0013] In this disclosure, "control" encompasses any measure that specifically influences the vehicle's behavior. Control is not limited to situations without a closed control loop. Rather, control can also occur within the framework of a regulation. Accordingly, the term "control information" includes any data that, directly or after processing steps, is capable of specifically influencing the vehicle's behavior. In some examples, the control information may contain specifications for vehicle behavior (e.g., driving parameters or trajectories to be observed). In other examples, control information may be received directly from the vehicle's actuators.

[0014] A second road user is considered obscured from the first road user if the first road user cannot detect the second road user in the given traffic situation. When this occurs can vary depending on the type and features of the first road user. For example, obscuration or detectability (also sometimes referred to as visibility) can depend on whether detection is carried out by one or more of the vehicle's sensors or by an occupant. In the first case, obscuration or detectability can depend on the type of sensor(s) and / or their arrangement.For example, there might be an uninterrupted line of sight between the sides of the first and second road users, but the second road user could still be obscured because a camera is positioned centrally on the front of the first road user, which cannot detect the second road user despite the uninterrupted line of sight between the sides. Obscuration or detectability can also be determined differently for sensors other than optical sensors (e.g., cameras) than by a line of sight.

[0015] In the present disclosure, "testing" means any verification of a function of a vehicle or vehicle component according to a predetermined specification. The term "testing" therefore encompasses both the verification and the validation of a vehicle or vehicle component.

[0016] A "vehicle" is any device for transporting passengers or goods. A vehicle may be autonomous or partially autonomous, but this need not be the case (e.g., assistance systems can be designed according to the techniques of this disclosure that provide a driver with information on the criticality of a traffic situation). In particular, a vehicle may be a passenger car or a truck. However, a vehicle may also be a (sub)sea craft, an aircraft, or a spacecraft. A vehicle may be a robot (or be described as such).

[0017] A "traffic situation" can be any situation in which the vehicle can be in operation. Typically, the vehicle is moving within a traffic situation. However, in some examples, a vehicle may also be stationary, at least temporarily, within a driving situation. Depending on the type of vehicle described above, the traffic situation can vary.

[0018] A "vehicle component" can be any element located within a system that contains a vehicle (i.e., is a vehicle). In some examples, a vehicle component might be located within the vehicle itself. In other examples, a vehicle component might be located outside the vehicle (but in communicative contact with the vehicle, e.g., a component in a backend or in the infrastructure through which the vehicle travels). A vehicle component can be a hardware component, a software component, or a component that contains both hardware and software (i.e., the specific function of the respective component is defined in hardware or software). An example of a vehicle component is a control unit for a vehicle function.

[0019] This disclosure covers all stages in the development and operation cycle of vehicles or vehicle components. Thus, the techniques described in this disclosure can be used in both the development and operation of a vehicle or vehicle component. Accordingly, the descriptions in this disclosure relate to vehicles and vehicle components at various stages of development and, consequently, in various forms. For example, a method for assessing the criticality of a traffic situation can be carried out using a simulation model of a vehicle or vehicle component. Therefore, whenever this disclosure refers to a vehicle or vehicle component, this description also relates to a vehicle or vehicle component in a model stage. Brief description of the characters Fig. Figure 1 is a flowchart illustrating various procedures according to the present disclosure. Fig. Figure 2a shows a first exemplary traffic situation according to the present disclosure. Fig. Figure 2b shows a second exemplary traffic situation according to the present disclosure. Fig. Figure 2c shows a third exemplary traffic situation according to the present disclosure. Fig. Figure 3 schematically shows an environment containing a vehicle in which the techniques of the present disclosure are used. Fig. Figure 4 schematically shows a test and / or development environment for vehicles or vehicle components in which the techniques of the present disclosure are used. Detailed description

[0020] Fig. Figure 1 is a flowchart illustrating various procedures according to the present disclosure. The middle column (II) shows a procedure for assessing the criticality of a traffic situation.

[0021] In a traffic situation, at least part of the surroundings of a first road user is obscured, and a second road user may be located within this obscured area (i.e., it is not necessarily apparent to the first road user whether or not a second road user is present in the obscured area). This part of the surroundings can be a section of a traffic area (e.g., a road, such as a multi-lane road).

[0022] The first road user is a vehicle (e.g., an autonomous vehicle, but in other examples also a semi-autonomous or assisted driving vehicle). In some examples, the second road user is also a vehicle. In other examples, however, the second road user can also be another object or subject participating in traffic (e.g., a pedestrian or an animal). In some examples, both the first and second road users are moving in the traffic situation (i.e., relative to a fixed frame of reference such as the Earth). However, the first and / or second road user can also be static, at least in phases, in the traffic situation (i.e., relative to a fixed frame of reference such as the Earth).

[0023] The procedure involves accessing a motion model for a traffic situation with a first road user and a second road user. The second road user is hidden from the first road user. The motion model models the future movement of the first and the second road user.

[0024] The motion model can be any model that allows predictions about the future motion states of the first and second road users. The motion model can, for example, model trajectories, but also individual parameters such as one or more distances between the first and second road users (e.g., a Euclidean distance, a longitudinal distance in the direction of travel or lane direction, or a lateral distance perpendicular to the direction of travel or lane direction). In some examples, the motion model models a minimum distance during the course of the traffic situation (e.g., a lateral or longitudinal minimum distance). The motion model can model individual parameters at specific points in time within the traffic situation. In some examples, the motion model can allow predictions about future motion states without requiring any further input data (e.g.,over a specific time horizon, such as the course of a particular traffic situation (e.g., up to 10 seconds into the future). In some examples, the motion model is based on an analysis of the accelerated motion of the first and second road users (e.g., assuming that the first and second road users are accelerating as rigid bodies). The motion model can be two-dimensional or one-dimensional.

[0025] In some examples, the motion model can be expressed by one or more equations (also called equations of motion) that allow statements about future motion states for the first and second road users. These equations can, for example, provide the statements described above, such as about trajectories or individual parameters like one or more distances between the first and second road users. The one or more equations can include the assumptions described below and / or measured or simulated input values ​​for one or more parameters as input variables.

[0026] Exemplary movement models and further aspects of the movement model are discussed below in connection with Fig. 2a - c discussed.

[0027] In some examples, the motion model may make assumptions about the position and / or speed (e.g., the initial speed in the traffic situation or a different instantaneous speed at a specific point in time) of the second road user. In some examples, an assumption may be that the second road user is positioned so that they are just out of the range of detection by the first road user (e.g., by a specific sensor of the vehicle and / or by an occupant; for example, it may be assumed that the second road user will move out of the obscured area immediately after the evaluated situation). Additionally or alternatively, the motion model may make assumptions about the speed (e.g., direction of movement and / or magnitude of speed) of the second road user.In some examples, the speed may be a maximum expected speed (e.g., a permissible maximum speed or a recommended speed, in some cases plus a certain amount).

[0028] Additionally or alternatively, the motion model can use assumptions about the behavior of the second road user (as input values). For example, one assumption could be that the second road user decelerates with a predetermined (variable or constant) deceleration or accelerates with a predetermined (variable or constant) acceleration. Additionally or alternatively, one assumption could be that the second road user performs a specific maneuver.

[0029] In some examples, the assumptions regarding the parameters and / or the behavior of the second road user can be selected taking into account a worst-case scenario or a reasonable worst-case scenario.

[0030] Additionally or alternatively, the motion model can use measured or simulated input values ​​for one or more parameters relating to the first road user. Unlike the second road user, the parameters for the first road user can be at least partially known to them. In some examples, the one or more parameters include the position of the first road user. Additionally or alternatively, the one or more parameters can include the speed of the first road user (e.g., the initial speed in the traffic situation or a different instantaneous speed at a specific point in time within the traffic situation). Furthermore, additionally or alternatively, the one or more parameters can include the distance of the vehicle to an object that obscures part of the surroundings.Additionally or alternatively, one or more parameters may include the speed and / or position of an object that obscures part of the environment.

[0031] In some examples, certain parameters for the first road user may also be determined through assumptions (and used by the motion model). This is particularly true for parameters that are not yet available at the time of the criticality assessment. Such parameters may include the reaction time, acceleration, and / or deceleration of the first road user.

[0032] In some examples, the motion model can use information about the positions of one or more sensors on the vehicle.

[0033] The procedure further includes processing the motion model to assess the criticality of the traffic situation. In some examples, processing the motion model may include evaluating a criticality metric that incorporates the future movements of the first and second road users as modeled by the motion model.

[0034] In some examples, the criticality metric can enable an assessment of criticality using a continuous criticality measure.

[0035] In other examples, the criticality metric can allow an assessment of criticality using three or more discrete levels of criticality (e.g., a non-critical level, a sub-critical level, and one or more critical levels, possibly with increasing seriousness).

[0036] The procedure also includes the assessment of the criticality of the traffic situation. The determined criticality can be used in various ways, as described below.

[0037] In some examples, the procedure may involve outputting a signal to a user interface based on the determined (and output) criticality of the traffic situation. For example, the output may be to an occupant (e.g., a driver) of the vehicle. The signal output may be via a suitable visual signaling device and / or audio signaling device and / or a tactile signaling device (e.g., a visual signaling element inside the vehicle). In some examples, a signaling device for outputting a signal based on the determined (and output) criticality of the traffic situation may be installed in the vehicle. In other examples, a signaling device for outputting a signal based on the determined (and output) criticality of the traffic situation may be carried in the vehicle (e.g., in an occupant's mobile device that may be communicatively linked to the vehicle).By sending a signal to a user interface based on the determined (and displayed) criticality of the traffic situation, a vehicle occupant (e.g., the driver) can be alerted to a critical situation and, if necessary, subsequently attempt to avert it by intervening in the vehicle's driving behavior. This can be beneficial for autonomous, semi-autonomous, and assisted driving vehicles alike.

[0038] Further possibilities for using the evaluation results obtained by means of the methods of this disclosure can be found in the testing of vehicles or vehicle components.

[0039] In some examples, a method for generating, selecting, and / or marking test data for a vehicle or vehicle component includes evaluating the criticality of traffic situations represented in the test data by a method according to one of the present disclosures and generating, selecting, and / or marking a specific set (e.g., subset) of the test data based on the determined criticality. For example, test data can be generated, selected, or marked in which a determined criticality is critical (e.g., above a predetermined criticality threshold or at a predetermined criticality level). In this way, the test data thus generated, selected, and / or marked can subsequently be used for testing vehicles or vehicle components.

[0040] In some examples, a method for generating, selecting, and / or marking training data for a vehicle or vehicle component includes evaluating the criticality of traffic situations represented in the training data by a method according to one of the present disclosures and generating, selecting, and / or marking a specific set (e.g., subset) of the training data based on the determined criticality. For example, training data can be generated, selected, or marked where a determined criticality is critical (e.g., above a predetermined criticality threshold or at a predetermined criticality level). In this way, the training data thus generated, selected, and / or marked can be used for training vehicles or vehicle components.

[0041] Additionally or alternatively, procedures for testing a vehicle or vehicle component may include setting up one or more test cases with traffic situations in which at least part of an environment is obscured from the vehicle (the first road user), and in which a second road user may be present. In some examples, test data for the test cases may be selected according to the techniques described above. In other examples, the setup may involve a test (simulated or in the field) in which the relevant traffic situations occur randomly or are deliberately induced.The procedure may further include determining whether the vehicle or vehicle component exhibits behavior in traffic situations such that its criticality, as defined in the present disclosure, remains below a certain criticality threshold (the criticality threshold may be a threshold value in the case of a continuous criticality metric or a threshold between two criticality levels in the case of a discrete criticality metric). If the vehicle or vehicle component exhibits behavior in traffic situations such that its criticality, as defined in the present disclosure, remains below a certain criticality threshold, the vehicle or vehicle component may, in some examples, be released (e.g., the vehicle or vehicle component may be manufactured and / or implemented).If the vehicle or vehicle component exhibits behavior in traffic situations such that its criticality, as defined in this disclosure, does not remain below the specified criticality threshold, the vehicle or vehicle component may not be released in some cases. In this case, the vehicle or vehicle component can be modified so that it subsequently exhibits behavior in traffic situations such that its criticality, as defined in this disclosure, remains below the specified criticality threshold.

[0042] In some examples, the techniques of this disclosure can be continuously performed during a test of a vehicle or vehicle component (e.g., in an endurance test). This allows for the determination of the extent to which, or how frequently, the vehicle or vehicle component causes critical situations. Based on the results of these investigations, a vehicle or vehicle component can then be evaluated (and, if necessary, released or modified).

[0043] The techniques of the present disclosure may, in some examples, help to reduce the effort required to test vehicles or vehicle components.

[0044] Additionally or alternatively, methods for training a vehicle or vehicle component may include training with traffic situations in which at least part of an environment is obscured from the vehicle (the first road user), and in which a second road user may be located, so that the vehicle or vehicle component exhibits behavior in the traffic situations such that a criticality according to the present disclosure remains below a certain criticality threshold (the criticality threshold may be a threshold value in the case of a continuous criticality metric or a threshold between two criticality levels in the case of a discrete criticality metric). In some examples, training data may be selected for training according to the techniques described above.

[0045] The present disclosure also relates to methods for generating control information for a vehicle in a traffic situation. In the left column (I) of Fig. Figure 1 shows an exemplary procedure for generating control information for a vehicle. The procedures for generating control information for a vehicle in a traffic situation can be carried out as an alternative to, or in addition to, the procedures for assessing the criticality of a traffic situation. On the one hand, in some examples, the criticality can be assessed (and, for example, signaled to an occupant) according to the present disclosure (and in particular as described above and below), and, in addition, a procedure for generating control information for a vehicle in a traffic situation can be carried out (e.g., to avoid a critical driving situation). In other cases, however, the procedures for generating control information for a vehicle in a traffic situation can be carried out without a procedure for assessing the criticality of a traffic situation. In some of these cases, for example,It is not necessary to explicitly evaluate a criticality metric.

[0046] In this traffic situation, at least part of the surroundings of a first road user is obscured, and a second road user may be present within this obscured area. Aspects and possible configurations of this traffic situation can be found above and below. The traffic situation is fundamentally no different from the examples with their assessment of criticality.

[0047] The procedure comprises accessing a motion model for a traffic situation with a first road user and a second road user, where the second road user is hidden from the first road user. The motion model models a future movement of the first and the second road user. The procedure further comprises processing the motion model to generate control information for the vehicle and outputting the control information for the vehicle or a vehicle component.

[0048] The motion models do not differ significantly from those used in the examples for assessing criticality. However, a criticality level can be specified to generate the control information, which the vehicle must not exceed (e.g., a criticality threshold, where the criticality threshold can be a threshold value in the case of a continuous criticality metric or a threshold between two criticality levels in the case of a discrete criticality metric).

[0049] In some examples, the motion model may be inverted or reformulated compared to the examples where criticality is assessed (e.g., one or more input values ​​may become output values ​​and vice versa). For example, one or more equations expressing the motion model for criticality assessment may be inverted or solved for a different parameter (e.g., the example equations of motion below). For example, an equation for assessing criticality may determine a distance, while an equation for determining control information may determine a velocity.

[0050] Based on the specified criticality, one or more parameters of the first vehicle can be calculated in some examples (these parameters could be, for example, one or more of the parameters measured or simulated in the criticality assessment examples). For instance, one or more distances between the first and second road users (e.g., a longitudinal distance in the direction of travel or a lateral distance perpendicular to the direction of travel) or a speed can be calculated (e.g., using one or more equations, possibly solved for another parameter, as described above or below). In some examples, maximum or minimum values ​​for one or more parameters can be derived from the specified criticality. For example, a maximum speed can be determined.Additionally or alternatively, a minimum value for a distance can be determined.

[0051] The control information for the vehicle or vehicle component can be determined from the calculated one or more parameters of the first vehicle. For example, the control information can be determined in such a way that, if the vehicle is controlled according to the control information, the one or more parameters are adhered to, not fallen below, or not exceeded (the latter, for example, if maximum or minimum values ​​are specified for the one or more parameters).

[0052] In some examples, the procedure may, in addition to carrying out the procedure for generating control information according to this disclosure 121, include controlling the vehicle using the generated control information. For example, the vehicle may be controlled so that its criticality remains below a predetermined criticality threshold. To control the vehicle, in some examples, instructions for one or more of the vehicle's systems that influence the vehicle's trajectory (e.g., a steering system, a braking system, an acceleration system, or other systems that, individually or in combination, determine the vehicle's trajectory) may be generated based on the control information.

[0053] In some examples, the procedures for generating control information for a vehicle can be executed continuously or intermittently during vehicle operation. Likewise, in some examples, the vehicle can be controlled continuously or intermittently using the generated control information.

[0054] The following will be based on the Fig. Sections 2a-2c explain some specific examples of the techniques of the present revelation.

[0055] In particular, in Fig. 2a - 2c show different traffic situations whose criticality can be assessed or for which vehicle control information can be calculated according to the techniques of this disclosure.

[0056] In some examples, the environment includes a traffic area in the form of a multi-lane road (for example, a highway or similar multi-lane road). Additionally or alternatively, the traffic situation may involve the vehicle overtaking another vehicle traveling in an adjacent lane, which obscures part of the traffic area. Alternatively, the traffic situation may involve the vehicle changing lanes and having another vehicle behind or in front of it that obscures part of the traffic area. Fig. Sections 2a-2c describe these traffic situations and partially reveal aspects of the present disclosure that are limited to these traffic areas and situations. However, the techniques otherwise described in the present disclosure are not limited to these types of traffic areas and situations. Rather, the techniques of the present disclosure can be applied to all traffic situations and areas in which a second road user may be obscured (and also some aspects in Fig. 2a - 2c can be applied to other traffic areas and other traffic situations. For example, traffic situations at intersections or traffic situations with oncoming traffic can be assessed in terms of their criticality, or control information can be calculated in such traffic situations.

[0057] Fig. Figure 2a shows a first exemplary traffic situation according to the present disclosure.

[0058] In this situation, on a traffic area in the form of a multi-lane road 21, there is a first vehicle 22 (first road user) and another vehicle 23, which obscures part of the traffic area for the first vehicle 22. Additionally, a second vehicle 24 may be located in the obscured area of ​​the traffic area in the form of a multi-lane road.

[0059] A motion model of the traffic situation can model the distance between the first and second vehicles. In some examples, the distance can be longitudinal (i.e., in the direction of travel). The distance can be a function of the speeds of the first and second vehicles 22, 24. Additionally or alternatively, the distance can be a function of the acceleration (e.g., a maximum acceptable acceleration) of the first vehicle 22 (e.g., in a first time window) and / or a minimum deceleration (e.g., a minimum acceptable deceleration) of the first vehicle 22 (e.g., in a second time window). Additionally or alternatively, the distance can be a function of the reaction times of the first and / or the second vehicle 22, 24. The motion model can be designed for the situation where the second vehicle 24 changes lanes into the lane of the first vehicle 22.Additionally or alternatively, the assumption may be included that the first vehicle 22 initially accelerates maximally for the duration of the first vehicle 22's reaction time and then decelerates with minimal deceleration. In some examples, the following relationship results for the longitudinal distance: dSLLocclusion=v1r+12a1,maxr2+(v1+a1,maxr)22b1,min−v222b2,max, where d SLLocclusion the longitudinal distance between the first and the second vehicle 22, 24, v1 an initial velocity of the first vehicle 22, v2 a velocity of the second vehicle 24, r a reaction time of the first vehicle 22, a 1, max a maximum acceleration of the first vehicle 22 at the beginning of the situation, b1, min; a minimum deceleration of the first vehicle 22 later in the situation and b 2, max a maximum delay of the second vehicle 24 seconds later in the situation.

[0060] In some examples, a criticality metric can be calculated based on an assumed distance between the first and the (possible) second vehicle 22, 24. This distance can be partially based on measurements (e.g., a measured distance of the first vehicle 22 to an object (e.g., the other vehicle 23) that causes the occlusion and / or a measured length of the object causing the occlusion). Additionally or alternatively, the distance can be partially based on assumptions (e.g., an assumed distance of the second vehicle 24 to an object causing the occlusion and / or an assumed length of the object causing the occlusion). The distance can be a longitudinal distance (i.e., the portion of the Euclidean distance in the direction of travel). In the situation of Fig. 2a A longitudinal distance can be calculated as: d=d1−ov+lov+dov−2 where d is the longitudinal distance, d 1-OVthe measured distance of the first vehicle to the further vehicle 23 (the obscuring object), I OV the measured or assumed length of the other vehicle 23 (the obscuring object) and d OV-2 the assumed distance between the second vehicle 24 and the further vehicle 23 (the obscuring object) are.

[0061] Based on the longitudinal distance according to the motion model and the assumed longitudinal distance between the first and the (possibly existing) second vehicle 22, 24, a criticality measure for the situation can be calculated according to a criticality metric.

[0062] In one example, the criticality metric is: SLLocclusion(d)=max(0,qq−1dSLLocclusiond−1q−1), where S LLocclusion the criticality measure and q is a normalization factor.

[0063] In other examples, instead of a longitudinal distance, a lateral distance or a Euclidean distance can be determined and used as the basis for calculating the criticality measure.

[0064] In some examples, a criticality assessment can only be made if a traffic situation is classified, according to one or more criteria, as one in which a second road user may be located in an obscured part of the first road user's surroundings. One of these criteria may, in some examples, be the extent of the obscured area. The assessment may take into account the speed of the first vehicle, the position of the object obscuring the area, and / or the distance between the first vehicle and the object obscuring the area.

[0065] In some examples, the determination of distances or the evaluation of the criterion for assessing criticality can take into account the line of sight of one or more sensors of the first vehicle 22 (e.g., based on the position and / or orientation of the corresponding sensor on the first vehicle 22). For example, a situation with occlusion (in which the techniques of the present disclosure are used) can be assumed to exist if a line of sight from a sensor of the first vehicle 22 to the second vehicle 24 is interrupted. The sensor of the first vehicle 22 can only detect the second vehicle 24 when there is an uninterrupted line of sight between the sensor and the side of the second vehicle 24 facing the first vehicle 22 (in the example of Fig. 2a a corner of the rear of the second vehicle 24 facing the first vehicle 22 is not (or no longer) obstructed by the further vehicle 23. If the line of sight is not obstructed in this situation, there is also no obstruction. The techniques of this disclosure are not applied in this case.

[0066] Fig. Figure 2b shows a second exemplary traffic situation according to the present disclosure.

[0067] In this situation, there is again a first vehicle 22 (first road user) and another vehicle 23 on a traffic area 21 in the form of a multi-lane road. The second vehicle obscures part of the traffic area 21 for the first vehicle 22. Additionally, a second vehicle 24 may be located in the obscured area of ​​the traffic area 21. However, the positions of the first and second vehicles 22, 24 are different from the first situation according to... Fig. 2a is reversed. A critical situation can arise if the first vehicle 22 changes lanes into the lane of the second vehicle 24, which may already be traveling there. The above statements therefore apply with the corresponding reversal of the roles of the first and second vehicles 22 and 24. For example, in the specific equation of motion for modeling the distance to Fig. 2a above the sizes for the first and for the second vehicle 22, 24 are swapped.

[0068] Furthermore, various assumptions can be made in these situations. For example, it can be assumed that the second vehicle, number 24, travels at a certain maximum speed.

[0069] Fig. Figure 2c shows a third exemplary traffic situation according to the present disclosure.

[0070] In this situation, on a traffic area 21 in the form of a multi-lane road, there is again a first vehicle 22 (first road user) and another vehicle 23, which partially obscures the traffic area 21 for the first vehicle 22. Additionally, a second vehicle 24 may be located in the obscured area of ​​the traffic area 21 in the form of a multi-lane road. The situation is similar to that in Fig. 2a, however, the other vehicle 23, which caused the obstruction, is traveling in the lane of the first vehicle 22. A critical situation can arise if the first vehicle 22 changes lanes into the lane of the second vehicle 24, which may be traveling there. The statements above therefore apply with appropriate adjustments.

[0071] In the situations described above, the first vehicle can avoid 22 critical situations by appropriately selecting the described parameters. This corresponds to generating control information as disclosed herein.

[0072] The present disclosure also relates to environments designed to execute the techniques of this disclosure. Since the techniques of this disclosure can be executed both in the development or testing of vehicles or vehicle components and in the operation of vehicles or vehicle components, the environments can also be configured differently. The environment can include a computing unit and corresponding storage device for executing the techniques of this disclosure. A computing unit according to this disclosure can be a stand-alone computing unit comprising one or more processors. In other examples, or additionally, a computing unit can be a distributed computing unit (e.g., across locations in or outside the vehicle, or multiple locations outside the vehicle).

[0073] In some examples, the environment includes (or is) a vehicle or vehicle component.

[0074] Fig. Figure 3 schematically shows an environment 30 with a vehicle 31 in which the techniques of this disclosure can be used. In some examples, the techniques of this disclosure for assessing the criticality of a traffic situation and / or for generating control information are executed in a computing unit 32 in the vehicle 31 (e.g., a control unit or a vehicle computer). Alternatively, the techniques of this disclosure for assessing the criticality of a traffic situation and / or for generating control information can be executed in a computing unit 33 outside the vehicle 31 (e.g., in a backend connected to the vehicle 31 via a communication link 34).In other examples, the techniques of the present disclosure can be carried out together to assess the criticality of a traffic situation and / or to generate control information in computing units outside 33 and inside 32 of the vehicle 31.

[0075] In other examples, the environment includes a test and / or development environment for vehicles or vehicle components. Fig. Figure 4 schematically shows a test and / or development environment 40 for vehicles or vehicle components, for which the techniques of the present disclosure are used.

[0076] The test and / or development environment 40 includes at least one computing unit 41. Furthermore, the test and / or development environment includes a vehicle or vehicle component 42. The vehicle or vehicle component 42 may be a model. In some examples, the test and / or development environment 40 is a simulation environment (i.e., the test and / or development environment simulates the vehicle or vehicle component). In other examples, the test and / or development environment 40 is a software-in-the-loop or hardware-in-the-loop environment. In still other examples, the test and / or development environment 40 is an environment for conducting field tests.

[0077] The present disclosure also relates to a computer program containing instructions which, when executed by a computing unit 32, 33, 41, cause the computing unit 32, 33, 41 to execute a procedure according to the present disclosure (e.g., to assess the criticality of a traffic situation or to generate tax information).

[0078] The present disclosure also relates to a computer-readable medium that stores a computer program according to the present disclosure (e.g., a solid-state storage device).

[0079] The present disclosure also relates to a signal that contains a computer program according to the present disclosure (e.g., a coded signal for an over-the-air update). QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 102023211082.2

[0002]

Claims

[1] Methods for assessing the criticality of a traffic situation, in the traffic situation, at least part of the surroundings (21) of a first road user (22) is obscured, and a second road user (24) may be located in the obscured part of the surroundings. where the first road user (22) is a vehicle, the procedure includes: Accessing (101) a motion model for a traffic situation with a first road user (22) and a second road user (24), where the second road user (24) is hidden from the first road user (22), where the motion model models a future movement of the first and second road users (22, 24), Processing (103) the movement model to assess the criticality of the traffic situation; and Expenditure (105) of the criticality of the traffic situation. [2] Method according to claim 1, wherein the processing comprises evaluating a criticality metric which incorporates the future movements of the first and second road users (22, 24) modeled by means of the motion model. [3] Method according to claim 2, wherein the criticality metric comprises an assessment of criticality using a continuous criticality measure or with three or more discrete criticality levels. [4] Method according to any one of claims 1 to 3, further comprising: Output of a signal (107) based on the determined criticality of the traffic situation to a user interface of the vehicle, in particular to an occupant of the vehicle. [5] Methods for selecting and / or marking test data for a vehicle or vehicle component, comprising: Assessment of the criticality of traffic situations represented in the test data by a method according to any one of claims 1 to 4; and Selecting and / or marking a specific subset of the test data based on the determined criticality. [6] Methods for testing a vehicle or a vehicle component, comprising: Setting up (109) one or more test cases with traffic situations in which at least part of an environment (21) is obscured for a first road user (22) in which a second road user (24) may be located; Determine (111) whether the vehicle or vehicle component exhibits behavior in traffic situations such that a criticality according to any of the preceding claims 1 to 4 remains below a certain threshold. [7] Methods for generating control information for a vehicle in a traffic situation, in the traffic situation, at least part of the surroundings (21) of a first road user (22) is obscured, and a second road user (24) may be located in the obscured part of the surroundings. where the first road user (22) is the vehicle, the procedure includes: Accessing (115) a motion model for a traffic situation with a first road user (22) and a second road user (24), wherein the second road user (24) is hidden from the first road user (22), where the motion model models a future movement of the first and second road users (22, 24); Processing (117) the motion model to generate control information for the vehicle and Output (119) of the vehicle tax information. [8] Methods for steering a vehicle, comprising: Performing the method for generating tax information according to claim 7; and Vehicle tax (121) using the generated tax information, in particular, where the vehicle is controlled in such a way that criticality remains below a certain threshold. [9] Method according to any one of claims 1 to 8, wherein the motion model models a future distance between the first road user (22) and the second road user (24), and wherein the motion model uses assumptions about the position and / or speed and / or acceleration of the second road user (24). [10] Method according to any one of claims 1 to 9, wherein the motion model uses measured input values ​​to one or more of: a position and / or speed and / or acceleration of the first road user (22); and / or a distance of the first road user (22) to an object (23), that obscures part of the surroundings (21); and / or a position and / or velocity and / or acceleration of the object (23) that obscures part of the environment (21). [11] Method according to any one of the preceding claims 1 to 10, wherein the method is carried out during the operation of the vehicle. [12] Environment (30; 40) designed to perform one of the methods according to any one of the preceding claims 1 to 11, in particular wherein the environment is a test and / or development environment for vehicles or vehicle components (40) and / or comprises a vehicle or vehicle component (30). [13] Computer program containing instructions which, when executed by a computing unit (32; 33; 41), cause the computing unit (32; 33; 41) to execute a method according to any one of claims 1 to 11. [14] Computer-readable medium or signal that stores and / or contains the computer program according to claim 13.

Images