Method and stopping distance determination device for determining a stopping distance to a vehicle in front when an ego vehicle comes to a stop in a stopping situation

Abstract

Method for determining a stopping distance (d) to be established between an ego-vehicle (1) and a front object, in particular a front vehicle (2), when an ego-vehicle (1) stops in a stopping situation, wherein the ego-vehicle is located behind the front object, in particular the front vehicle, in the stopping situation viewed in the direction of travel, wherein at least one stopping assistance function of the ego-vehicle (1) is provided with a stopping parameter characteristic of the stopping distance (d) to be established for its execution, wherein a stopping situation parameter (AS) characteristic of the stopping situation of the ego-vehicle (1) is taken into account when determining the stopping parameter.According to the invention, the at least one stopping parameter is determined as a function of a swarm stopping parameter, which is determined on the basis of swarm data generated by a swarm of vehicles in stopping situations characterized by the at least one stopping situation parameter (TS).

Description

[0001] The present invention relates to a method and a stopping distance determination device for determining the stopping distance to a vehicle in front or to an object in front when an ego-vehicle comes to a stop in a stopping situation. The present invention further relates to a method for determining the distance behavior of a swarm of vehicles when stopping behind an object in front, preferably a vehicle in front, depending on the distance situations of the vehicles in the swarm.

[0002] Many production vehicles use Adaptive Cruise Control (ACC) to regulate the distance to the vehicle in front. These functions also offer the option of the vehicle braking automatically to a complete stop. This increases comfort, for example in city traffic or traffic jams.

[0003] To offer the driver a more natural / human-like driving experience, the chosen stopping distance is an important factor in addition to vehicle deceleration. Both too large and too small a distance can create an unpleasant driving experience.

[0004] In previous implementations, the stopping distance was determined by the developer (expert model). However, the optimal stopping distance from a developer's perspective may not correspond to the expectations of an average customer and, moreover, is not adjusted based on the situation (e.g., in traffic jams, rain, darkness, slippery conditions, etc.). A situation-dependent adjustment of the stopping distance, however, may be what the average customer expects. For example, the optimal distance in a traffic jam on the highway may differ from the optimal distance in city traffic.

[0005] The literature contains approaches that consider factors influencing the distance maintained between two vehicles in (partially) automated longitudinal control (e.g., a larger safety distance in bad weather). However, no explicit methodology is described that extracts desired behavior from real-world data and automatically / partially incorporates this into a higher-dimensional model, or that parameterizes this model without expert knowledge. The higher the dimension of the descriptors to be considered, the more complex their identification and representation in an expert model becomes. Therefore, the existing literature typically considers only a few dimensions, whose influence is only partially coupled.

[0006] From US patent 2012 / 0176234 A1, a system called "Adaptive Cruise-with-Braking" (ACB) is known for controlling a vehicle's braking response distance (BRD). This system monitors a variety of trigger conditions (e.g., environmental parameters). The BRD is adjusted in response to detected trigger events or conditions, such as system fault conditions, ABS / traction / stability events, road surface conditions, traffic conditions like congestion / density, and current and / or recent speeds of vehicles in the same or adjacent lanes.

[0007] US patent 2013 / 0289844 A1 discloses a smart cruise control system that includes a braking distance calculation unit to determine the relative speed and distance between vehicles in order to calculate the braking distance. Furthermore, a road surface detector is provided on the opposite side of the vehicle to detect the current road surface condition. The braking distance calculation unit adjusts the braking distance according to the current road surface condition.

[0008] US Patent 2013 / 0317719 A1 proposes a control system and a procedure for stopping a vehicle that reduces the jerk when a vehicle is stopped without driver intervention. It determines whether there is a distance sufficient to allow the controlled vehicle to come to a smooth stop over the set target stopping distance.

[0009] If the interrelationships within the assistance function are not adequately addressed, customers may repeatedly override the function, resulting in reduced trust and acceptance. For example, under adverse conditions (e.g., rain and darkness), the stopping distance may be set too short and (subjectively) perceived as a risk (even if no collision would occur). This could lead to the permanent deactivation of the function or, in future generations of the assistance system, to its removal altogether, potentially resulting in financial losses.

[0010] The present invention is based on the objective of overcoming the disadvantages known from the prior art and providing a means of determining the (stopping) distance to be set when an ego vehicle stops behind a front object, such as a vehicle driving ahead, or a distance determination device which provides a more human-like or natural distance selection adapted to the respective stopping situation.

[0011] The object of the invention is achieved by the subject matter of the independent claims. Advantageous embodiments and further developments of the invention are the subject matter of the dependent claims.

[0012] In a method according to the invention for determining a stopping distance to be established between the ego-vehicle and a front object, in particular a front vehicle, when an ego-vehicle stops in a stopping situation (of the ego-vehicle), the ego-vehicle is located behind the front object, in particular the front vehicle, in the stopping situation (of the ego-vehicle) viewed in the direction of travel (of the ego-vehicle).

[0013] The stopping distance is understood in particular to be the distance between the ego vehicle and the object or vehicle in front, which is established or exists when the ego vehicle is stationary, i.e., after the stopping process of the ego vehicle has been completed.

[0014] The object in front could be a road user, such as a vehicle ahead or a pedestrian. It could also be a construction vehicle or another mobile object. Furthermore, it could be a stationary object located (at least temporarily) on the roadway and / or the (planned) trajectory of the ego-vehicle, particularly in front of the ego-vehicle when viewed in its direction of travel.

[0015] Preferably, the ego vehicle is the next vehicle in relation to the front object, in particular the front vehicle, so that in particular no further vehicle is located between the ego vehicle and the front object or the front vehicle.

[0016] In this scenario, the object in front, particularly the vehicle in front, is preferably already stationary at the beginning of the stopping situation (of the ego vehicle). It is possible that the vehicle in front was initially moving in the same direction as the ego vehicle and then braked to a standstill – for example, at an intersection or in another traffic situation.

[0017] The start of a stopping situation can be understood as the point in time at which the ego-vehicle determines that it needs to brake, specifically to a standstill. The start of a stopping situation can also be understood as the point in time at which the ego-vehicle initiates and / or begins a stopping process.

[0018] It is also conceivable that the vehicle in front is still moving at the beginning of the stopping situation (of the ego vehicle), but preferably is already braking and, even more likely, has come to a complete stop. For example, the vehicle in front could be approaching an intersection (such as one with a red light) or an obstacle and is therefore braking.

[0019] In this process, at least one stopping assistance function of the ego vehicle is provided with a stopping parameter characteristic of the stopping distance to be set. This stopping parameter could be, for example, the intended stopping distance that should be established between the ego vehicle and the object in front when stationary, i.e., after the stopping process of the ego vehicle has been completed.

[0020] Preferably, at least one stopping assistance function is performed, at least partially automatically and preferably fully automatically, depending on the stopping size.

[0021] Preferably, the stopping assistance function is a function that brakes the ego-vehicle to a standstill. Particularly preferably, the stopping assistance function is a function of the ego-vehicle that brakes the ego-vehicle to a standstill, depending on the object or vehicle in front, in such a way that it comes to a stop at the required distance from the object or vehicle in front.

[0022] It is also conceivable that the stopping assistance function is a warning and / or notification function, by means of which a warning message and / or notification is issued to the driver of the ego vehicle depending on the stopping distance (visual and / or audible). For example, the driver could be warned of an impending loss of stopping distance if a predetermined distance to the object in front is not reached and / or if the ego vehicle is decelerating too slowly.

[0023] In determining the stopping quantity, (at least) one stopping situation quantity characteristic of the stopping situation of the ego vehicle, preferably determined by the ego vehicle, is taken into account.

[0024] The stopping situation may involve a traffic situation and / or a weather situation and / or a road surface situation (coefficient of friction, for example prevailing slipperiness of the road) and / or a brightness situation and / or a visibility situation (such as limited visibility in fog, heavy rain) and / or a vehicle type situation (such as a truck as the vehicle in front, so that there is limited visibility of the traffic behind it).

[0025] The stopping situation parameters are, in particular, parameters detectable by the ego-vehicle that are suitable for characterizing a stopping situation. A stopping situation can be characterized by a single stopping situation parameter. Preferably, a stopping situation is characterized by a multitude of stopping situation parameters (such as traffic density, occupied lane, amount of rainfall, time of day, risk of slipperiness, and the like).

[0026] Preferably, the stopping situation parameter(s) are detected by at least one sensor device and / or environmental sensing device of the ego vehicle and / or determined based on (raw) sensor data acquired by the at least one sensor device and / or environmental sensing device. However, it is also conceivable that at least one stopping situation parameter is determined based on (or received by) communication data received by a communication device of the ego vehicle. The communication device can be a vehicle-to-X communication device (V2X communication device), in particular a vehicle-to-vehicle communication device.

[0027] According to the invention, the at least one stopping parameter is determined as a function of a swarm stopping parameter, which is determined on the basis of swarm data generated by a swarm of vehicles in stopping situations characterized by the at least one stopping situation parameter.

[0028] The stopping distance is preferably determined by means of a stopping distance determination device (described in more detail below), in particular a processor-based device, which is especially preferably a (permanent) component of the Ego vehicle.

[0029] Stopping situations characterized by at least one stopping factor (preferably by a multitude of stopping factor) are understood to mean, in particular, that although the stopping situations may differ from one another, they all share the same at least one stopping factor (preferably in the multitude of stopping factor). The stopping situations may, for example, be recorded in geographically different areas that are similar in one characteristic, such as a specific weather condition (e.g., freezing rain) or a specific road surface condition (e.g., gravel road).

[0030] This advantageously achieves that, for the specific stopping situation in which the ego-vehicle is currently located, swarm data is taken into account to determine the stopping distance of the ego-vehicle. This data reflects the stopping behavior of a swarm of vehicles in comparable stopping situations. By using a swarm stopping parameter to determine at least one stopping parameter of the ego-vehicle, it is advantageously possible to determine a typical or average stopping behavior, or an average stopping distance tailored to the specific stopping situation of the ego-vehicle.

[0031] In other words, the stopping distance is determined taking into account real fleet data, which reflects the current stopping situation of the ego vehicle.

[0032] The proposed method offers the advantage that it allows for a situationally adaptable selection of a (desired) stopping distance, where evaluation of swarm data has shown that users feel comfortable with this setting in relation to the vehicle in front.

[0033] In a preferred method, the swarm stopping value is characteristic of a stopping distance of the vehicle swarm to the respective object in front (given a stopping situation of the respective swarm vehicle characterized by at least one stopping situation parameter). Advantageously, the respective stopping distance between the vehicles in the swarm and their respective objects in front (from the swarm data) is determined in their respective stopping situations, and an (approximately average) swarm stopping value is derived from these. The swarm stopping value is preferably characteristic of a uniform distance behavior or stopping behavior of the vehicle swarm in the stopping situation characterized by the at least one stopping situation parameter.

[0034] In a further preferred method, the determination of the stopping distance takes into account a multitude of stopping situation parameters determined by the ego vehicle, which are characteristic of the stopping situation of the ego vehicle. The swarm stopping distance is characteristic of a stopping distance maintained by a swarm of vehicles in stopping situations characterized by the multitude of stopping situation parameters. This advantageously makes it possible to use several stopping distance parameters to define or specify a given stopping situation. This advantageously achieves a more precise differentiation of the stopping situation, so that a more precise consideration of the influencing factors on the choice of a stopping distance is advantageously attainable.

[0035] In a further preferred method, the at least one stopping situation parameter and preferably the plurality of stopping situation parameters are selected from a group of stopping situation parameters which includes traffic density, a lane in use, a number of lanes, an amount of rain, a time of day (day / night), a risk of slipperiness, a road category (urban, rural, motorway), a type of the object in front and preferably of the vehicle driving ahead (e.g. truck, car, two-wheeler), a selected driving profile, a driver type, a coefficient of friction, a road surface, a speed limit or parameters characteristic of these, as well as combinations thereof.

[0036] Additionally or alternatively, the group may include a quantity characteristic of a brightness situation, ambient brightness, illumination, lighting, season and / or temperature, as well as combinations thereof or with the quantities mentioned above.

[0037] Preferably, at least three, preferably at least four, preferably at least five, preferably at least seven and especially preferably at least ten stopping situation parameters are used to characterize the stopping situation of the ego vehicle.

[0038] It is conceivable that different stopping situations require a different number of stopping situation parameters for their characterization. For example, a single stopping situation parameter may suffice for a stopping situation parameter that is very dominant with regard to the choice of stopping distance (such as the presence of an extremely slippery road surface). For another stopping situation, a multitude (e.g., more than four) stopping situation parameters may be necessary for its characterization (and differentiation from other stopping situations with different stopping distance choices) (e.g., urban area; intersection topology; type of vehicle in front; ambient light level; traffic density).

[0039] Preferably, the ego vehicle has a storage device, in particular a vehicle-bound one, on which a large number of different stopping situations or the stopping situation parameters required for them (or their assumed values ​​or value ranges when the respective stopping situation is present) as well as the swarm stopping parameters assigned to the various stopping situations (at least one each) are stored.

[0040] Preferably, the ego-vehicle or the stopping distance determination device first determines at least one characteristic stopping situation parameter to (more precisely) characterize the current stopping situation of the ego-vehicle. For example, based on the multitude of stopping situations stored on the memory device, the ego-vehicle can select exactly one stopping situation (by comparing the values ​​or value ranges of the stopping situation parameters currently assumed by the ego-vehicle with the stored values ​​or value ranges of the stopping situation parameters) based on the data collected by the ego-vehicle from the at least one sensor device and / or environment detection device and / or communication device.

[0041] In a further preferred method, a selection of distance situation variables, in particular groups or clusters of distance situation variables, and preferably a plurality of clusters of distance situation variables, is specified, depending on which the swarm stopping variable and / or the stopping variable is determined.

[0042] Preferably, at least ten different, preferably at least 15 different, preferably at least 25 different and particularly preferably at least 50 different stopping situations (or their respective distance situation sizes) as well as their respective assigned swarm stopping size are stored on the storage device of the Ego vehicle.

[0043] Preferably, an update of the stopping situations and / or the swarm stopping parameters assigned to each of them can be implemented. It is also conceivable that the stopping situations could be supplemented or expanded to include new stopping situations.

[0044] The present invention is further directed to a method, in particular a computer-implemented method, for determining the spacing behavior of a swarm of vehicles when stopping behind a (respective) front object, preferably a front vehicle, depending on the spacing situations of the vehicles in the swarm, comprising: - Providing swarm data generated in distance situations of the vehicle swarm, comprising a (predefined) multitude of distance situation parameters characteristic of the respective distance situation and at least one distance parameter characteristic of the distance of the respective swarm vehicle to the respective front object. - Determining a selection of distance situation parameters from the multitude of distance situation parameters based on the swarm data in such a way that the selected distance situation parameters are characteristic of a respective (specific and / or uniform) or the respective (uniform) distance behavior of the vehicle swarm.

[0045] A specific and / or uniform spacing behavior of the vehicle swarm preferably refers to the selection of an equal spacing range (e.g. in 0.5 m increments) and / or distance (value) to the (respective) vehicle in front.

[0046] In other words, distance situation parameters are identified that allow for a statistically valid prediction of a desired distance behavior (such as the target stopping distance to a vehicle or object in front) of a user of an ego-vehicle in a given stopping situation, which can be described or characterized according to these distance situation parameters. The statistical relevance and / or validity is assessed based on the swarm data of the vehicle swarm.

[0047] This offers the advantage that distance situations which are similar to each other with regard to the selected distance situation parameters and which validly and reproducibly lead to a comparable distance behavior of the vehicle swarm can be identified and characterized (and defined) via the distance situation parameters.

[0048] If the presence of values ​​for these identified distance situation parameters is detected in an ego vehicle, it can be deduced that a distance situation exists which leads to a distance behavior of the respective users known from the swarm data. In this way, many different distance situations can be identified based on the swarm data, all of which lead to a uniform distance behavior of the users. For each uniform distance behavior (e.g., target distance to the object or vehicle in front), at least one swarm distance parameter is preferably determined, which is characteristic of the uniform distance behavior (e.g., target distance to the object or vehicle in front) of the vehicle swarm.

[0049] These distance situation parameters, which identify the respective distance situations, are preferably provided to the ego vehicles together with the swarm distance parameter, which characterizes the uniform distance behavior.

[0050] If one of these distance situations is then identified at an ego vehicle (based on the values ​​for the distance situation parameters recorded by the ego vehicle), the distance of the ego vehicle to the front vehicle or front object is preferably determined using the swarm distance parameter provided to the ego vehicle (e.g. on an ego vehicle-bound storage device).

[0051] To determine the exact swarm spacing in a stopping situation, data from a swarm of vehicles is preferably considered, comprising at least 50 vehicles, preferably at least 100 vehicles, and most preferably at least 1000 vehicles. This allows the spacing and stopping behavior of at least 50 (different) vehicles in a single stopping situation to be taken into account.

[0052] Swarm data is preferably determined or generated as follows: Preferably, when using swarm data to determine the swarm distance size, only data from swarm vehicles that brake to a standstill are considered (especially as an initial trigger).

[0053] Preferably, when using swarm data to determine the swarm distance size, only data from swarm vehicles are considered where a front object or a front vehicle is located in front of the swarm vehicle (especially as a further or first boundary condition).

[0054] Preferably, for the purpose of determining the swarm spacing, only data from vehicles in the swarm are considered where the stopping time is preferably between 5 and 200 seconds, preferably between 10 and 120 seconds (and particularly preferably between 10 and 100 seconds) (especially as a further and preferably as a second boundary condition). This allows, for example, parking situations to be excluded. Preferably, only data from vehicles in the swarm are considered for the purpose of determining the swarm spacing where the stopping time is greater than 1 second, preferably greater than 2 seconds, preferably greater than 5 seconds, and particularly preferably greater than 10 seconds.When determining the swarm distance size, only data from swarm vehicles with a stopping time of less than 250 seconds, preferably less than 120 seconds and particularly preferably less than 50 seconds are considered in the swarm data.

[0055] Preferably, when determining the swarm distance size, only data from swarm vehicles are considered in which no longitudinal assistance system is active and / or in which the swarm vehicle is operated manually (especially as a further and preferably as a third boundary condition).

[0056] Preferably, time series for the stopping situation variables (for a relevant or specified time period) are generated (as swarm data).

[0057] Preferably, time series data can be recorded and / or used to generate the swarm data used to determine the stopping distance for each stopping situation parameter(s), covering a predetermined time period before the stopping period of the respective swarm vehicle. Preferably, the time series for each stopping situation parameter(s) extends from the beginning of the stopping situation of the respective swarm vehicle or from the beginning of the stopping process of the respective swarm vehicle (until the stopping period of the swarm vehicle). Preferably, the time series for each stopping situation parameter includes values ​​of the respective stopping situation parameter(s) at predetermined intervals (e.g., at regular intervals). This allows the temporal dependence of various influencing factors and their effect on the selection of the stopping distance to the vehicle or object in front to be taken into account.

[0058] Preferably, time series data (as swarm data) are generated for the distance values ​​(for a relevant or specified time period).

[0059] In a preferred method (for evaluating the swarm data), a user can specify a selection of distance situation parameters, and the specified selection of distance situation parameters is checked on the basis of the swarm data to determine whether it is characteristic of a uniform (or comparable and / or specified and / or specifiable) distance behavior of the vehicle swarm.

[0060] Preferably, the swarm data is used to check whether the user-specified selection of distance situation parameters (and / or their values) leads to a statistically significant uniform distance behavior of the users of the swarm vehicles.

[0061] In a further preferred approach, the selection of distance situation parameters is determined using pattern recognition methods. Data analysis methods such as unsupervised learning, pattern recognition techniques, and clustering methods can be employed. For example, patterns in the dataset can be grouped, resulting in a (uniform) distance behavior (e.g., via clustering). This process is preferably automated, and higher-dimensional dependencies, particularly those involving at least five, and preferably at least ten, distance situation parameters, are preferably captured. The number of clusters can be estimated or specified by experts or determined automatically, for example, using the elbow criterion.

[0062] A uniform spacing behavior can exist, for example, if the distance maintained between the respective swarm vehicle and the object or vehicle in front is within the same predetermined distance range.

[0063] For this purpose, different distance ranges can be predefined, to which the respective distance maintained by a swarm vehicle is assigned. A specific (discrete) number of distance ranges can be predefined and / or predefinable for categorizing the distance behavior.

[0064] Preferably, a predetermined number of value ranges (categories) are specified for each distance situation parameter. Preferably, each stopping situation, which is characterized with respect to the distance situation parameter, is assigned (exactly) one of these value ranges (categories) (of the respective distance situation parameter).

[0065] Preferably, clusters are identified or determined, whereby the clusters specify a selection of distance situation variables and preferably at least one range of values ​​of the respective distance situation variable, which (with a given probability) lead to (exactly) one distance range (uniform distance behavior).

[0066] The present invention is further directed to a stopping distance determination device, in particular a processor-based device, for an ego-vehicle to determine a stopping distance to be established when the ego-vehicle stops in a stopping situation between the ego-vehicle and a front object, in particular a front vehicle, wherein the ego-vehicle is located behind the front object, in particular the front vehicle, in the stopping situation when viewed in the direction of travel.

[0067] The stopping distance determination device is suitable and intended to provide at least one stopping assistance function of the Ego vehicle with a stopping parameter characteristic of the stopping distance to be set, and the stopping distance determination device is suitable and intended to take into account a stopping situation parameter characteristic of the stopping situation of the Ego vehicle when determining the stopping parameter.

[0068] According to the invention, the stopping distance determination device is suitable and intended to determine the at least one stopping parameter as a function of a swarm stopping parameter, which is determined on the basis of swarm data generated by a swarm of vehicles in stopping situations characterized by the at least one stopping situation parameter.

[0069] It is therefore also proposed within the framework of the stopping distance determination device according to the invention that real fleet data be taken into account when determining a target stopping distance.

[0070] Preferably, the stopping distance determination device is configured, suitable, and / or designed to execute the above-described method for determining the stopping distance between the ego-vehicle and a front object when it comes to a stop in a stopping situation, as well as all the process steps already described above in connection with the method, either individually or in combination. Conversely, the method can be equipped with all the features described within the stopping distance determination device, either individually or in combination.

[0071] The present invention further relates to a vehicle, in particular a motor vehicle, comprising a stopping distance detection device described above for a vehicle according to one embodiment. The vehicle may in particular be a (motorized) road vehicle.

[0072] A vehicle can be a motor vehicle, which is in particular a driver-operated vehicle ("driver only"), a semi-autonomous vehicle, an autonomous vehicle (for example, at autonomy levels 3, 4, or 5 (according to the SAE J3016 standard)), or a self-driving vehicle. Autonomy level 5 refers to fully automated vehicles. The vehicle can also be a driverless transport system. In this case, the vehicle can be driven by a driver or drive autonomously. Furthermore, in addition to a road vehicle, the vehicle can also be an air taxi, an aircraft, or another means of transport or vehicle type, such as an aircraft, watercraft, or rail vehicle.

[0073] The present invention further relates to a computer program or computer program product, comprising program means, in particular a program code, which represents or encodes at least some and preferably all of the process steps of the two (above described) methods according to the invention and preferably one of their described preferred embodiments and is designed for execution by a processor device.

[0074] The present invention further relates to a data storage device on which at least one embodiment of the computer program according to the invention or a preferred embodiment of the computer program is stored.

[0075] Further advantages and embodiments can be seen from the attached drawings: It shows: Fig. 1 a schematic representation of an ego vehicle with a stopping distance determination device according to the invention in a stopping situation; and Fig. 2a, Fig. 2b Illustrations for identifying clusters from given indicative situational variables.

[0076] Fig. Figure 1 shows a schematic representation of an ego vehicle 1 with a stopping distance determination device 10 according to the invention in a stopping situation, which here is given by an intersection situation in which the ego vehicle is brought to a standstill behind a front vehicle 2, here another car, at a traffic light by means of an at least partially automatic driver assistance system.

[0077] The reference sign d denotes a desired distance between the ego vehicle 1 and the front vehicle 2, which is to be achieved by means of the at least partially automatic driver assistance system when the ego vehicle is stationary.

[0078] Reference numeral 10 designates a stopping distance determination device of the ego vehicle, in particular a processor-based device, which determines a desired stopping distance (or a characteristic stopping distance value) depending on the currently given stopping situation, in particular detected by at least one sensor device of the ego vehicle.

[0079] The stopping distance detection system of the Ego vehicle determines the stopping distance or stopping distance size in such a way as to offer a more natural or human-like driving experience.

[0080] This is done by evaluating the spacing behavior of swarm data or fleet data.

[0081] The analysis of ACC usage patterns and real driving situations has shown that drivers choose different stopping distances to the vehicle in front under the influence of different environmental conditions.

[0082] The two Fig. 2a and Fig. 2b illustrates an example of the proposed analysis of swarm data or evaluation of distance behavior in fleet data.

[0083] This study primarily considers scenarios in which a driver manually brakes to a standstill behind another vehicle. The selected stopping distance d can be enriched in a dataset with vehicle-internal measurement parameters (labels). Examples of labels used include: - Traffic density - Lane in use / Number of lanes - Rainfall amount - Time of day (day / night) - Risk of slippery conditions - Road category (urban, rural, motorway) - Type of vehicle in front (e.g. truck or car) - selected driving profile - Driver type - Coefficient of friction - Road surface - Speed ​​limits

[0084] This dataset can be used to develop a situation-specific driver assistance function that takes all these influencing factors into account and offers the driver a more natural / human-like driving experience. The data-driven approach, which is described in more detail in the following steps, offers the advantage that characteristics of different descriptors or stopping distance situation parameters can be linked together to form high-dimensional information that can be (partially) automatically weighted in relation to the stopping distance d measured in real-world data.

[0085] For example: the stopping distance varies greatly depending on the vehicle. a) Rain, urban environment, cobblestones and a car as the target object (or foreground object) compared to: b) Rain, motorway, asphalt and a truck as the target object (or foreground object).

[0086] These dependencies do not need to be described, implemented, and validated by experts in the approach described below.

[0087] The approach can be described as follows in the context of the stopping distance: Example methodology: 1. Data acquisition and preprocessing

[0088] Scenarios or stopping situations are recorded from fleet data or swarm data of a vehicle swarm, preferably fulfilling the following criteria: - Initial trigger t_0: EGO vehicle brakes to a standstill - Boundary conditions 1: Object / vehicle in front of the EGO vehicle - Boundary conditions 2: n seconds < t_{standing} < m seconds (e.g. n=10 and m = 120) - Boundary conditions 3: No longitudinal assistance system active (manual driving) - Generating time series for the relevant time period that contain the descriptors or stopping situation variables described above. 2. Data analysis / grouping

[0089] The aim of the data evaluation or grouping (of the analyte situation variables or descriptors) is to extract (stopping) situations / scenarios from which a comparable distance results after the EGO vehicle has stopped compared to the vehicle in front.

[0090] This is preferably done in the following steps: Firstly, the distance between the Ego vehicle and the front object 2 (front vehicle) at rest is preferably divided into groups (in a first step). For example, the division into groups (a) e.g. small, medium, large distance or (b) in classes that differ, for example, in predetermined distances, such as 0.5m increments, take place.

[0091] In this step, distance categories are preferentially created or predefined. Each distance is preferentially assigned to at least one of a large number of (predefined) distance categories.

[0092] In a subsequent (preferably second) step, the time series compiled in preprocessing are preferably assigned to the distance categories.

[0093] In a subsequent (preferably third) step, the descriptors or indicative situational variables in these sub-datasets are analyzed. The aim is to find patterns that reproducibly lead to a desired distance behavior (i.e., the distance behavior obtained through a given distance category, for example, the distance category "small distance").

[0094] This can be done, for example, through experts who formulate a hypothesis and validate it based on the data: e.g., the experts can formulate the hypothesis: "a large distance is left behind a truck" (situations are less visible).

[0095] The disadvantage is that higher-dimensional relationships are difficult to grasp.

[0096] Additionally or alternatively, the analysis can preferably be performed automatically, for example, using unsupervised learning. Patterns in the dataset can be grouped, resulting in distance behavior (e.g., via clustering). This process can be automated and also capture higher-dimensional dependencies. The number of clusters can be estimated by experts or determined automatically (e.g., using the elbow criterion).

[0097] Each category is divided into subcategories, for example into 4 subcategories (clusters, e.g. via k-means): This means, for example, that for the preferred variant chosen above with distance classes, four (or more / less) clusters are formed for each of the distance categories "small, medium, large" across all descriptors or stopping situation variables.

[0098] The result is an assignment of the respective distance category in variations of the value ranges that the descriptors or stopping situation variables span.

[0099] The two Fig. 2a and Fig. 2b illustrates this with an example of four descriptors or four stopping situation variables and four clusters as a heat map (depending on point density) for the (distance) category “Small distance” ( Fig. 2a) and the (distance) category “Large distance”.

[0100] The two tables each use dark grayscale values ​​to show that the respective value range or category of the stopping situation variable is very dominant in the cluster, while the light grayscale values ​​are not dominant in the cluster.

[0101] The stopping situation variables or descriptors chosen in this example are “rainfall amount”, “traffic density”, “vehicle type” and “time of day”.

[0102] Each stopping situation variable (ASG or descriptor) is further subdivided into categories. For example, the stopping situation variable (ASG or descriptor) "Rainfall" is subdivided into the four categories "none," "low," "medium," and "high." Similarly, the stopping situation variable (ASG or descriptor) "Traffic Density" is subdivided into the three categories "low," "medium," and "high," each marked with the reference symbol ASK. Finally, the stopping situation variable (ASG or descriptor) "Vehicle Type" is subdivided into the two categories ASK "Car" and "Truck." Finally, the stopping situation variable (ASG or descriptor) is subdivided into the two categories ASK "Day" and "Night."

[0103] For each of the two distance classifications “Small distance” ( Fig. 2a) and “Large distance” ( Fig. 2b) then results, for example, in the four different clusters C given, which highlight dominant categories ASK of various given stopping situation variables.

[0104] If one considers this in Fig. In the example given in 2a of the cluster C (1st cluster) specified in the first line, the data analysis of the swarm data of a vehicle swarm has shown, for example, that the probability of choosing a small distance is high when the stopping situation is "no rain and medium traffic density".

[0105] According to the information in the bottom line of the Fig. As a further example from the clusters specified in 2a (“4th cluster”), the probability that a small distance is chosen is high when a stopping situation is characterized by the stopping situation variables “no rain”, “front vehicle: car”, “time of day: day”.

[0106] For example, it has been found that for a large distance ( Fig. 2b) In cluster C obtained from the first row, only the stopping situation variable AS "amount of rainfall" has a dominant classification, namely the classification ASK "a lot". Therefore, according to the swarm data, "a lot" of rain alone already leads to a large distance.

[0107] This approach can be used to illustrate which descriptors are dominant in which distance categories and how the respective feature dependencies are. For subsequent model building, it must be ensured that no combination of cluster and value combination of descriptors occurs twice within the higher-level categories (in this case, an expert-based determination of a higher-level category would be necessary). 3. Model building:

[0108] In model building, the corresponding problem can be viewed from an inverted perspective. This means that, in the final step of model building, the existing descriptors define a distance category, and the derived relationships are transferred into a model. Depending on the descriptors present in real-world operation and the desired stopping distance determined from data, this model can transmit a desired target distance for braking to a standstill to the assistance system, which varies depending on the situation.

[0109] For example (analogous to the example above, see Fig. 2a) Small target distance at standstill, for example when: - No rain, medium traffic density - High traffic density - Light / moderate rain, medium to high traffic density, daytime - No rain, car as target object, day

[0110] Large distance when: - Lots of rain - Low / Medium traffic, trucks as target object - Light / moderate rain, light traffic, night - Moderate / heavy rain, light traffic, trucks

[0111] The applicant reserves the right to claim all features disclosed in the application documents as essential to the invention, provided they are novel individually or in combination compared to the prior art. It is further noted that the individual figures also describe features which may be advantageous on their own. A person skilled in the art will immediately recognize that a particular feature described in a figure may be advantageous even without incorporating other features from that figure. Furthermore, a person skilled in the art will recognize that advantages may also arise from a combination of several features shown in individual or different figures. Reference symbol list 1 Ego vehicle 2 Front vehicle 10 Stopping distance determination device d stopping distance AS stopping situation size ASK classification of a stopping situation variable C Cluster 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] US 2012 / 0176234 A1

[0006] US 2013 / 0289844 A1

[0007] US 2013 / 0317719 A1

[0008]

Claims

[1] Method for determining a stopping distance (d) to be established between an ego vehicle (1) and a front object, in particular a front vehicle (2), when an ego vehicle (1) stops in a stopping situation, wherein the ego vehicle is located behind the front object, in particular the front vehicle, in the stopping situation in the direction of travel, wherein at least one stopping assistance function of the ego vehicle (1) is provided with a stopping parameter characteristic of the stopping distance (d) to be established for its execution, wherein a stopping situation parameter (AS) characteristic of the stopping situation of the ego vehicle (1) is taken into account when determining the stopping parameter. characterized by, that at least one stopping parameter is determined as a function of a swarm stopping parameter, which is determined on the basis of swarm data generated by a swarm of vehicles in stopping situations characterized by the at least one stopping situation parameter (TS). [2] Method according to claim 1, characterized by , that the swarm stopping size is characteristic of a stopping distance of the vehicle swarm to the respective object in front. [3] Method according to any one of the preceding claims, characterized by , that in determining the stopping quantity a multitude of stopping situation quantities determined by the ego vehicle (1), characteristic for the stopping situation of the ego vehicle (1), are taken into account, and that the swarm stopping quantity is characteristic for a stopping distance (d) taken by a swarm of vehicles in stopping situations characterized by the multitude of stopping situation quantities. [4] Method according to any of the preceding claims, characterized by , that at least one stopping situation parameter and preferably the multitude of stopping situation parameters is selected from a group of stopping situation parameters which includes traffic density, a lane in use, a number of lanes, an amount of rain, a time of day (day / night), a risk of slipperiness, a road category (urban, rural, motorway), a type of the object in front and preferably of the vehicle driving ahead (e.g. truck, car, two-wheeler), a selected driving profile, a driver type, a coefficient of friction, a road surface, a speed limit or parameters characteristic of these, as well as combinations thereof. [5] Method according to any one of the preceding claims, characterized by , that a large number of clusters of distance situation variables are specified, depending on which the swarm stopping size and / or the stopping size is determined. [6] Method for determining the spacing behavior of a swarm of vehicles when stopping behind a front object, preferably a front vehicle, depending on the spacing situations of the vehicles in the swarm, comprising: - Providing swarm data generated in distance situations of the vehicle swarm, comprising a multitude of distance situation parameters characteristic of the respective distance situation and at least one distance parameter characteristic of the distance of the respective swarm vehicle to the respective front object. - Determining a selection of distance situation parameters from the multitude of distance situation parameters based on the swarm data such that the selected distance situation parameters are characteristic of the distance behavior of the vehicle swarm. [7] Method according to any of the preceding claims, characterized by, that a user can specify a selection of distance situation parameters and the specified selection of distance situation parameters is checked on the basis of the swarm data to see if it is characteristic of an identical and / or specified and / or specified distance behavior of the vehicle swarm. [8] Method according to any of the preceding claims, characterized by that the selection of distance situation parameters is determined using pattern recognition methods. [9] Stopping distance determination device (10) for an ego vehicle (1) for determining a stopping distance (d) to be established when the ego vehicle (1) stops in a stopping situation between the ego vehicle (1) and a front object, in particular a front vehicle (2), wherein the ego vehicle is located behind the front object, in particular the front vehicle, in the stopping situation viewed in the direction of travel, wherein the stopping distance determination device (10) is suitable and intended to provide at least one stopping assistance function of the ego vehicle (1) with a stopping parameter characteristic for the stopping distance (d) to be established, wherein the stopping distance determination device is suitable and intended to take into account a stopping situation parameter (AS) characteristic for the stopping situation of the ego vehicle (1) when determining the stopping parameter. characterized bythat the stopping distance determination device is suitable and intended to determine at least one stopping parameter as a function of a swarm stopping parameter, which is determined on the basis of swarm data generated by a swarm of vehicles in stopping situations characterized by the at least one stopping situation parameter (TS). [10] Vehicle (1), in particular motor vehicle, comprising a stopping distance detection device (10) according to the preceding claim.

Images