Method and system for identifying a following driving situation

Abstract

Method (3a, 3b, 3c) for adapting automatic longitudinal guidance of a vehicle to driver behavior comprises determining (31) by means of one or more predetermined criteria whether a following situation exists in which the vehicle follows a target object at a following distance intended by a driver of the vehicle under the influence of the operating behavior of the driver. If it is determined (31) here that a following situation exists, an indication can be triggered (32a) to the driver in order to suggest an automatic driving function to the driver. Alternatively or additionally, an actual following distance detected in the following situation is used within the scope of a learning algorithm for learning (32b) the following distance for the automatic longitudinal guidance of the vehicle and / or for setting (32c) the following distance within the scope of the automatic longitudinal guidance of the vehicle.

Description

Technical Field

[0001] This invention relates to a computer-implemented method for identifying following vehicle situations, a processing apparatus, a computer program for performing this method, and a computer-readable storage medium storing such a computer program. Furthermore, this invention also relates to a system for identifying following vehicle situations and a vehicle equipped with such a system. Background Technology

[0002] Many modern motor vehicles are equipped with autonomous driving functions, particularly those enabling automatic longitudinal guidance of the vehicle. A widely used example of an autonomous driving function is automatic distance-speed adjustment, often referred to as distance-adjustable cruise control or adaptive cruise control (ACC). Here, if the vehicle ahead is traveling slower than a target speed set for the vehicle, the following distance is adjusted to the target distance using automatic engine and braking intervention. The target following distance can be, for example, a specific second interval, meaning it can be preset to a specific time interval in which a vehicle equipped with ACC should follow the vehicle ahead.

[0003] Because different drivers have different following distance preferences, the target following distance can usually be set by the driver within certain boundaries. For example, it can be stipulated that the driver can globally set the following distance at multiple distance levels, where a corresponding distance-speed characteristic curve is stored for each distance level. However, such predefined distance levels only conditionally reflect individual needs and habits. While they reflect the average of ordinary drivers, they do not offer the possibility of adapting to specific driver preferences.

[0004] In conventional systems, the preset distance-speed characteristic curves follow a fixed pattern. At low speeds, the distance per second is typically high. The higher the speed, the lower the corresponding distance. However, this doesn't necessarily correspond to the driver's expected behavior. In dense urban traffic, drivers might, for example, prefer dense driving and choose longer distances on highways and rural roads to easily reach their destination. These preferences are not reflected in the conventional distance levels because they don't correspond to the average driver. Therefore, due to the limitation of only a few settable distance levels, only a limited number of individual driver preferences can be considered.

[0005] Furthermore, without considering different driving conditions, such as city traffic or highways, drivers must manually switch between distance levels in different situations. This necessity of manual operation can affect the comfort experience and, in addition, distract the driver from their driving task.

[0006] In the prior art, several solutions are already known to learn desired distances from the driver’s driving behavior and to take these desired distances into account in automated longitudinal guidance.

[0007] For example, DE 10 2015 016 993 A1 describes a method for acquiring driving characteristics of a vehicle’s driving assistance system, comprising the steps of: recording distance measurement signals from the vehicle’s distance measurement system and driving dynamic measurement signals of the vehicle in an inactive driving assistance system. The storage device stores recorded distance measurement signals and driving dynamic measurement signals; driving behavior is obtained from the stored distance measurement signals and driving dynamic measurement signals, wherein the distance and / or time distance to the vehicle ahead is obtained from the distance measurement signals; and the obtained driving behavior is stored. Furthermore, a driving assistance adjustment method is proposed, particularly an ACC adjustment method using this method, and a driving assistance system for performing the driving assistance adjustment method.

[0008] In addition, it has been suggested to use neural networks to learn individualized driver behavior related to following distance (see, for example, HUANG, Xiuling; SUN, Jie; SUN, Jian, A car-following model considering asymmetric driving behavior based on long short-term memory neural networks. Transportation research part C: emerging technologies, 2018, 95. Jg., S.346-362.).

[0009] There is a problem with learning following distance during driving periods controlled or at least influenced by the driver: not every distance set by the driver to the vehicle ahead actually corresponds to its typical expected distance for following. Road traffic is dynamic. New vehicles may cut in, vehicles may accelerate, new speed limits may be introduced, or sudden braking may be required. In all these situations, the driver does not intentionally maintain a constant distance behind the target object. Instead, the distance can vary significantly and temporarily exceed or fall below the actual expected distance. Summary of the Invention

[0010] The objective of this invention is to describe a method and system for recognizing following conditions and utilizing this knowledge, particularly in conjunction with longitudinal guidance, in autonomous driving functions.

[0011] This task is addressed by the subject matter of the independent claims. Advantageous embodiments are described in the dependent claims.

[0012] A first aspect of the invention relates to a computer-implemented method for identifying following situations. This method should particularly identify scenarios where a vehicle is in a following situation anticipated and influenced by the vehicle driver.

[0013] According to some implementation variations described in further detail thereafter, this also relates to a method for adapting the automatic longitudinal guidance of a vehicle to driver behavior detected in following situations.

[0014] Vehicles can be, in particular, motor vehicles. Here, the term "motor vehicle" is understood specifically as a land vehicle that is moved by mechanical force and is not connected to rails. In this sense, a motor vehicle can be constructed, for example, as a passenger car, a motorcycle, or a tractor-trailer.

[0015] The vehicle may be equipped with autonomous driving features for longitudinal guidance, such as ACC (Adaptive Cruise Control) or more comprehensive features, such as supporting combined longitudinal and lateral guidance.

[0016] The term "autonomous driving function" as used within the scope of this document generally refers to a vehicle function capable of achieving autonomous driving. Here, "autonomous driving" is understood as driving with automatic longitudinal and / or lateral guidance. This could be, for example, driving for extended periods on a highway or driving within a time limit while parked. The term "autonomous driving" includes autonomous driving with any degree of automation. Exemplary levels of automation are Driver Assistance, Semi-Autonomous Driving, Conditional Automated Driving, High Automated Driving, and Full Automated Driving (each with an increasing degree of automation). The five automation levels mentioned above correspond to SAE Levels 1 through 5 according to the standard SAE J3016 (SAE - Society of Automotive Engineers) as of April 30, 2021. In Driver Assistance (SAE Level 1), the system performs longitudinal or lateral guidance in specific driving situations, with the driver expected to assume all remaining aspects of the dynamic driving task. In Semi-Autonomous Driving (SAE Level 2), the system assumes longitudinal and lateral guidance in specific driving situations, where the driver must continuously monitor the system as in Driver Assistance. In Conditional Automated Driving (SAE Level 3), the system provides longitudinal and lateral guidance under specific driving conditions, without requiring continuous driver monitoring; however, the driver can guide the vehicle for a certain period as requested by the system. In Highly Automated Driving (SAE Level 4), the system guides the vehicle under specific driving conditions even when the driver does not respond to intervention requests, thus the driver is no longer considered a backup. In Fully Automated Driving (SAE Level 5), the system performs all aspects of dynamic driving tasks under any road and environmental conditions, and is also controlled by a human driver.

[0017] In the steps of the method, one or more predetermined criteria are used to determine whether a following situation exists, in which a vehicle, influenced by the driver's actions, follows a target object (especially the vehicle in front) at a following distance anticipated by the driver. Here, it is not necessary to determine with absolute safety that the vehicle follows the target object at the following distance anticipated by the driver. Rather, the aim is to determine whether specific criteria, further illustrated below, are met, which, based on experience, suggest that this is likely the case.

[0018] Therefore, what is being identified is when the driver (may) intentionally adjusts a specific following distance to the target object. As a prerequisite, one can first check, for example, whether "manual" longitudinal guidance (e.g., implemented in a conventional manner by manipulating the accelerator and / or brake pedals) is currently in effect, or whether an automated driving function with longitudinal guidance, such as ACC, is not activated.

[0019] In this case, the activated speed limiter should only be considered an activated longitudinal guidance driving function if the activated speed limiter is currently restricting the vehicle's speed, because the driver manually applies longitudinal guidance when the speed is below the speed limiter's limit.

[0020] When such following situations are confirmed, it can be assumed with a certain probability that the actual following distance reflects the driver's desired driving behavior, and is therefore particularly important for the purpose of learning individualized expected following distances.

[0021] Accordingly, according to the implementation variant, in another step of the method, the actual following distance detected in the following situation is used within the range of the learning algorithm to learn the following distance for automatic longitudinal guidance.

[0022] Learning algorithms can adapt distance adjustment to individualized driving behavior by learning specific distance characteristic curves. The goal here is to store the driver's preferences and adjust the distance to match their expectations. Therefore, distance adjustment should be adapted to the driver's individual driving characteristics in order to design distance adjustment for the driver's comfort as much as possible.

[0023] Specifically, this can be achieved by automatically adapting the stored distance-speed characteristic curve to the driver's driving behavior observed in the identified following situation, which is used to adjust the distance to the vehicle ahead according to speed within the range of automatic longitudinal guidance.

[0024] The distance-speed characteristic curve can be designed such that it assigns the target following distance to multiple speed values ​​(referred to as the support point speed within the scope of this specification). The target following distance can be, for example, a time distance (hereinafter sometimes referred to as the second distance), which indicates at what time the vehicle should follow the vehicle ahead. If the second distance is, for example, 1.5 seconds, then the longitudinal guidance of the vehicle is adjusted so that the vehicle arrives at the observation point (i.e., for example, longitudinally along the shared lane of the vehicle and the vehicle ahead) 1.5 seconds after the vehicle ahead. Alternatively, the target following distance can also be described as an actual distance, i.e., an actual distance in units of length.

[0025] The adaptation of the characteristic curve can be performed during a defined learning phase, within a following situation controlled or at least influenced by the driver. In other words, a prerequisite for the learning phase may be the determination of the following situation, as described with reference to the previous method steps. If this prerequisite is given, then the learning phase can occur, and the detected distance-speed characteristic curve can be adapted within the scope of the learning algorithm based on the actual following distance detected in the identified following situation. Specifically, within the scope of the learning algorithm, it can be stipulated that if a following situation determined according to the above method steps does not currently exist, then the learning phase does not occur.

[0026] It can also be stipulated that if all other conditions used to determine the hypothetical following situation are met, then a predetermined anti-shake time is waited before the learning phase begins, i.e., before the learning algorithm is actually activated. Therefore, the probability that the actual following distance processed within the range of the learning algorithm corresponds to the stably set expected following distance can be further increased. The anti-shake time can be, for example, in the range of 1 to 10 seconds, such as 3 seconds.

[0027] To adapt to the distance-speed characteristic curve, for example, a method can be implemented, the subject of which is the subject of a patent application filed by the applicant with the German Patent and Trademark Office on January 15, 2024, entitled "Method for Learning Following Distance and System for Automatic Longitudinal Guiding of Vehicles." Therefore, the content of that patent application is incorporated into this patent application. The method for identifying following conditions described in this patent application can be advantageously used to define one or more learning stages of the method for learning following distance according to the aforementioned other patent applications. In particular, the condition for initiating the learning stage may be determining the following conditions.

[0028] According to the implementation, when the learning phase of the learning algorithm used to learn the following distance is activated, the driver is informed by means of an appropriate output device, such as an optical display or an audio output device. For example, the driver may additionally be informed of the current "direction" in which the learned following distance is being adapted, i.e., whether the driver's current driving behavior will cause (e.g., with respect to the current speed) an increase or decrease in the previously stored target following distance.

[0029] The knowledge about the existence of following vehicles obtained in the above-described steps can also be advantageously used in addition to algorithms for learning following distance.

[0030] Therefore, according to an alternative implementation variant (which may be implemented alternatively or additionally relative to the previously described implementation variant related to the learning algorithm), an instruction is triggered to the driver when it is determined that a following vehicle situation exists.

[0031] Here, the condition "when it is determined that a following situation exists" is not necessarily to be understood as a sufficient condition for triggering the instruction. In other words, according to some embodiments, when a following situation is identified in the above method steps, it is not necessary to trigger the instruction under all circumstances. Instead, it may (but not necessarily) be stipulated that additional prerequisites must be added to trigger the instruction. According to some embodiments, the identification of the method situation according to the above method steps can be regarded as a necessary prerequisite for triggering the instruction. Generally, the description of triggering the instruction when a following situation is determined should be understood as outputting the instruction based on the determination of the following situation, that is, considering the situation.

[0032] This instruction can have the function of suggesting autonomous driving functions to the driver when following another vehicle. In identified following situations, longitudinally guided autonomous driving functions, such as ACC (Adaptive Cruise Control), can alleviate the driver's task of adjusting the appropriate following distance. Proactively providing this function can improve driver comfort because it allows for more frequent and targeted activation of automatic longitudinal guidance in appropriate driving scenarios.

[0033] The instruction can be conveyed to the driver via appropriate output devices, such as visual, auditory, and / or tactile means. Furthermore, it is conceivable to use a human-machine interface (HMI) designed to detect driver input, for example, by touching an input field on a touchscreen that is displayed along with the instruction, to activate the autonomous driving function in response to the instruction.

[0034] According to another implementation variation, the actual following distance detected during following maneuvers is used to set the following distance within the range of the vehicle's automatic longitudinal guidance. This additional implementation variation can be implemented alternatively or additionally to each of the two aforementioned implementation variations (regarding the use of the actual following distance for learning algorithms or for instructing the driver). Therefore, in addition to the influence on the following distance learned by the learning algorithm, the longitudinal guidance-based autonomous driving function can be configured to (furthermore directly) react to the actual following distance detected during the identified following maneuver.

[0035] If an automated driving function with automatic longitudinal guidance is activated after determining the following situation, the following distance to the target vehicle can be set, for example, based on the actual following distance detected (at least temporarily) deviating from the target following distance during the following situation. This target following distance is typically set for the automated driving function at relevant speeds based on a stored distance-speed characteristic curve. Within the scope of the automated driving function, the following distance can be set, for example, in a way that gradually approaches the target following distance. This is particularly meaningful when the actual following distance detected during the following situation deviates significantly from the target following distance according to the characteristic curve. For example, the driver can initially intentionally maintain a relatively large following distance from the vehicle ahead without activating ACC. When the driver subsequently activates ACC, it is beneficial for driver comfort to adapt the longitudinal adjustment so that the ACC function does not suddenly shorten the following distance to the target following distance, but rather the following distance traversed by ACC only slowly approaches the target following distance. Crucially, however, the system must identify in advance the following situation where the driver intentionally sets a larger following distance.

[0036] Regarding the predetermined criteria used to determine the presence of following vehicles, several possible implementations are subsequently described, which can be combined with each other. As explained more precisely below, these criteria may, for example, involve the vehicle's own driving dynamics parameters (e.g., acceleration or speed) and / or (also) parameters related to the target object (e.g., the speed difference between the vehicle ahead and the vehicle itself). Such parameters may, for example, be provided by the vehicle's data bus system (possibly in conjunction with an environmental sensor system). Furthermore, special driving situations, such as approaching a traffic light or intersection, may be considered, and in particular, evaluated as indicators of the absence of following vehicles.

[0037] Typically, these criteria are based on the idea that only (e.g., with a relatively constant following distance) stable following scenarios should preferably be identified as following scenarios in the sense of this method, in which it can be assumed that the driver intentionally adjusts their desired following distance and their primary objective is not other traffic conditions. These situations can be considered important, for example, for the purpose of learning the driver's desired following distance, making it meaningful to perform a learning phase when identifying these following scenarios. In contrast, dynamic traffic conditions or situations where the selected distance deviates unreasonably and significantly from the behavior learned so far should preferably be kept out of consideration, i.e., not identified as following scenarios in the sense of this method. The learning phase of the algorithm for learning following distance can, for example, be specifically set outside of these less important driving situations.

[0038] According to the implementation method, as a condition for determining the existence of following traffic (e.g., necessary and / or sufficient), it is checked whether the detected actual following distance is within a predetermined value range.

[0039] Actual following distance can be considered, for example, in the form of seconds. Here, the condition for determining following distance can be that the second distance is not less than a predetermined minimum value, which can be, for example, less than or equal to 0.5 seconds. Alternatively or additionally, the condition for determining following distance can be that the second distance is not greater than a predetermined maximum value, where the maximum value can be, for example, greater than or equal to 3 seconds. This is based on the idea that the second distance to the vehicle ahead must move within a realistic range for the driver's intended following distance. For example, second distances less than 0.5 seconds and greater than 3 seconds can be considered unrealistic in this sense, and special cases deviating from typical intentional following distances should be indicated where necessary.

[0040] According to another embodiment that can be combined with the above-described embodiments, as a condition for determining the existence of following traffic (e.g., necessary and / or sufficient), it is checked whether the vehicle's acceleration and / or the target object's acceleration are within a predetermined value range. Here, the value range may be, for example, a relatively small interval near 0, such as -0.5 m / s². 2 Up to +0.5m / s 2 The range. In other words, the condition used to determine following distance can be that the acceleration in absolute value does not exceed a predetermined range, for example, 0.5 m / s². 2 The maximum value. This can be achieved by checking if the vehicle acceleration is near 0, considering only fixed following conditions. This is based on the idea that in dynamic situations, distance adjustment is usually not the primary adjustment target, and therefore the actual distance detected is less important.

[0041] Additional criteria, which can be considered alternatives to or supplements to the aforementioned criteria, relate to the speed difference with the target object. A prerequisite for classifying a driving situation as following another vehicle may be that the speed difference is not too large, such that the actual following distance remains relatively constant. Therefore, according to an embodiment, as a (e.g., necessary and / or sufficient) condition for determining the existence of following another vehicle, it can be checked whether the absolute value of the difference between the speed of the target object and the speed of the vehicle (or vice versa) is less than a predetermined maximum speed difference. Alternatively, as a prerequisite for determining the existence of following another vehicle, it can be checked whether the absolute value of the difference between the speed of the target object and the speed of the vehicle is less than or equal to a predetermined maximum speed difference.

[0042] Within the scope of this invention, determining whether a following situation exists can also be based on whether a special driving situation is identified. In particular, it can be stipulated that if a special driving situation exists, then it is determined that a following situation does not exist (or at least it is uncertain whether a following situation exists). In other words, specific driving situations can be selectively excluded, for example, for the purpose of learning following distance.

[0043] Special driving situations may involve road direction, (especially) previous traffic guidance, or (especially previous) traffic guidance infrastructure, as illustrated in specific examples below.

[0044] Depending on possible implementation variations, special driving situations include approaching traffic lights (especially red or yellow ones). That is, a constant driving phase cannot usually be assumed when approaching a traffic light. Here, depending on the driver type, driving behavior varies between coasting and braking shortly before the traffic light. The driver's primary focus is no longer the expected distance to the vehicle ahead, but rather the traffic light itself.

[0045] In another implementation variant, special driving situations include vehicles approaching intersections, especially intersections where vehicles must respect the right-of-way of other vehicles that may be present.

[0046] Studies of actual driving have shown that drivers typically back up slightly before intersections and traffic lights, thus temporarily increasing the distance to the vehicle ahead beyond the actual expected distance set by the driver while following another vehicle. Furthermore, following distance can be reduced or increased by the driver of the vehicle ahead braking or accelerating at intersections or traffic lights, for example, to take advantage of the yellow phase of the traffic light.

[0047] It is also possible that special driving situations involve an upcoming turn of the vehicle. When approaching a turn, it is assumed that the driver reduces the following distance due to necessary upcoming braking, such that the resulting following distance can no longer be assumed to represent the normal expected distance when following another vehicle. The turn can be identified, for example, by means of a selected navigation route and / or by means of the detected manipulation of the driving direction indicator (especially of the vehicle itself, but optionally of the vehicle in front).

[0048] Regarding the specific driving situations exemplified above, particularly information concerning approach to traffic lights, intersections, and / or turns, this information can be provided, for example, by digital maps combined with the vehicle's location, for example, from a global navigation satellite system. Alternatively or additionally, this information can also be provided in a known manner by means of the vehicle's environmental sensor system, which may include one or more cameras.

[0049] According to another embodiment that can be combined with the above-described embodiments, as a prerequisite for determining the existence of following traffic, it is checked whether the actual following distance detected for the vehicle's speed deviates from the following distance set for that speed based on the stored distance-speed characteristic curve by no more than a predetermined size.

[0050] For example, it can be checked whether the actual following distance is within a predetermined tolerance zone around the stored (and if necessary, learned over a certain period of time) distance-speed characteristic curve if the actual following distance is entered into the distance-speed chart for the corresponding speed.

[0051] The tolerance band, or more generally the predetermined size, can be correlated with the consistency of driving behavior as represented in the stored distance-speed characteristic curve. The consistency of the stored driving behavior can be quantified as a reasonableness value. This reasonableness value can be calculated, for example, by incrementing the integral count during the learning phase. The longer the learning phase takes place near the learned characteristic curve, the higher the reasonableness.

[0052] Based on this reasonableness value, a range can be defined around the stored values ​​that still allows for learning. The smaller the reasonableness value, the larger this range becomes. Ranges outside this range are excluded because they deviate too far from normal driver behavior and may therefore indicate special circumstances.

[0053] Here, the present invention is based on the idea that the longer the actual distance is learned in the relevant distance range during the learning process, the more reliable the learned following distance can be considered. A long learning phase (or multiple learning phases totaling a long time) can, for example, result in a driver typically maintaining a following distance in the range of 1.3s to 1.4s at a specific speed. If a significant increase in the actual following distance is determined in some situation (i.e., a following distance significantly greater than 1.4 seconds), then this may be an unexpected following situation, i.e., a situation where the vehicle in front is traveling relatively far ahead of the vehicle at approximately the same speed, and the driver of the vehicle has not intentionally followed that vehicle or adjusted the expected following distance. Such outliers are identified by comparing the actual following distance with the characteristic curve learned so far, taking into account reasonable values, and differing from the (actual) following situation. Therefore, for example, it can be avoided that further learning of the characteristic curve due to outliers within the actual following distance traveled can lead to a deterioration of the characteristic curve.

[0054] Based on the above description, in the implementation, the predetermined size is set according to a reasonableness value, wherein the reasonableness value is increased or decreased by comparing the difference between the actual following distance detected at the corresponding time point and the target following distance applicable according to the distance-speed characteristic curve at the corresponding time point at multiple previous time points with the last observed time point. In particular, it can be specified that the larger the reasonableness value, the smaller the predetermined size is set (and vice versa). This can be achieved, for example, by having the predetermined size (i.e., the width of the tolerance band around the characteristic curve, as described above) determined by a mathematical expression in which at least one term is inversely proportional to the reasonableness value k or the reasonableness value is subtracted from a constant.

[0055] Using the plausibility values ​​further elaborated below, we can make the following statement regarding the reliability of the currently learned distance. Based on this, we can suppress the learning of erroneous distances that deviate significantly from the distances learned so far.

[0056] Reasonableness values ​​can be calculated for the entire characteristic curve. However, it is also possible to perform calculations individually for each or all support points of the characteristic curve in order to make the distinction more accurate.

[0057] Besides using a reasonableness value to determine the presence of following distance, as mentioned above, the reasonableness value can also be used in other ways within the scope of learning algorithms used to learn following distance. For example, the patent application "Method for Learning Following Distance and System for Automatic Longitudinal Guiding of Vehicles" dated January 15, 2024, mentioned at the beginning, describes how, within the scope of such a learning algorithm, a reasonableness factor can be defined based on the reasonableness value when determining a new target following distance, such that the reasonableness factor decreases as the reasonableness value increases and conversely increases as the reasonableness value decreases. If the reasonableness value is very high, the adaptation of the characteristic curve to the corresponding actual following distance will be reduced or slowed down. Conversely, if the reasonableness value is low, a clearer adaptation of the characteristic curve to the current actual value can be performed, so that the learned characteristic curve changes more quickly across multiple calculation steps.

[0058] The change in the target object is determined according to the implementation of the method proposed herein. Subsequently, as a prerequisite (e.g., necessary and / or sufficient) for determining the existence of a following situation, it is checked whether a predetermined minimum duration has elapsed since the change in the target object was determined. The change in the target object can be caused, for example, by another vehicle inserting itself between the vehicle and the previous vehicle ahead (the previous target object) and thus becoming the new target object. This corresponds to the experience that the driver of the vehicle dynamically needs a certain amount of time based on the situation to adjust their desired distance from the new target object. Therefore, it may be meaningful to wait for a certain minimum duration before (possibly, i.e., when other possible prerequisites are also met) to determine the following situation again. The minimum duration may, for example, be in the range of 2 to 12 seconds.

[0059] According to the improved scheme, the minimum duration can be dynamically set based on parameters representing the situation, such as specifically the speed of the vehicle and / or the new target object and / or whether the vehicle is in an environment within or outside the city. Therefore, the more dynamic the situation, the longer the time lag (i.e., the minimum duration) after the target object changes.

[0060] A second aspect of the invention is a (data) processing apparatus having at least one processor and designed to execute the method according to a third aspect of the invention by means of the at least one processor. According to some embodiments, this may also involve processing apparatuses spatially distributed (e.g., on multiple mutually spaced processors or microcontrollers). The processing apparatus may, for example, be a control apparatus (or a portion thereof), such as a control apparatus for controlling the automatic longitudinal guidance of a vehicle. However, embodiments in which the processing apparatus is arranged outside the vehicle are also possible. For example, it is conceivable that the processing apparatus is part of a back-end server to which the vehicle is connected via a wireless communication path.

[0061] A third aspect of the present invention is a system for identifying following vehicle situations.

[0062] The system includes a processing device according to the second aspect of the invention, or is connected or connectable to such a processing device (e.g., located on or at the rear of a vehicle) via data technology. Therefore, the system can use the method for identifying following situations according to the first aspect of the invention. Thus, the above and below description of the method according to the invention and its possible designs can be understood as similar to, and vice versa, the system according to the invention.

[0063] The system is designed to output instructions to the driver via an output device when the processing device determines that a following vehicle is in motion, so as to suggest the automatic driving function to the driver.

[0064] Alternatively or additionally, the system may be designed to perform automatic longitudinal guidance of the vehicle by setting the following distance to the target object based on the actual following distance detected in the identified following driving situation.

[0065] For example, it can be specified that the system or the system's control module (e.g., control device) uses a distance-velocity characteristic curve that has been learned or updated by means of a method as the basis for distance adjustment.

[0066] Within the scope of this invention, the system can also be designed to set a following distance to a target object based on the actual following distance detected (at least temporarily) as deviating from the target following distance when an autonomous driving function with automatic longitudinal guidance of the vehicle is activated after a following situation is determined, the target following distance being generally applicable to the autonomous driving function at the relevant speed according to a stored distance-speed characteristic curve.

[0067] The system, for example, can set the following distance within the scope of the autonomous driving function in a way that gradually approaches the target following distance. This is particularly meaningful when the actual following distance detected during following maneuvers deviates significantly from the target following distance according to the characteristic curve. For example, the driver can initially intentionally maintain a relatively large following distance from the vehicle in front without activating ACC. When the driver subsequently activates ACC, the system can adapt the longitudinal adjustment so that the following distance is not immediately shortened to the target following distance, but rather gradually approaches the target following distance through the following distance traveled by ACC.

[0068] A fourth aspect of the invention is a vehicle, particularly a motor vehicle, having a system according to a third aspect of the invention.

[0069] A fifth aspect of the invention is a computer program comprising commands that, when executed by a processing device (e.g., a processing device according to the second aspect of the invention), cause the processing device to perform a method according to the first aspect of the invention. Here, the computer program can be divided into multiple independent subroutines, each of which can be executed by different processing devices (e.g., multiple independent processors) that are spatially distant from each other if necessary. Therefore, a processing device according to the second aspect of the invention can be designed, and in particular programmed, to execute the computer program according to the fifth aspect of the invention.

[0070] A sixth aspect of the invention is a computer-readable storage medium comprising commands that, when executed by a (distributed if necessary) processing device, cause the processing device to perform a method according to a first aspect of the invention. In other words, a computer program according to a fifth aspect of the invention can be stored on the computer-readable storage medium. Attached Figure Description

[0071] The present invention will now be described in detail with reference to the embodiments and the accompanying drawings.

[0072] Figure 1 An exemplary and illustrative system for identifying following vehicles is shown.

[0073] Figures 2A to 2C The steps of a method for identifying following vehicle situations are illustrated, both by example and schematic.

[0074] Figure 3A The distance-velocity characteristic curve, as well as the upper and lower boundary characteristic curves, are shown as examples.

[0075] Figure 3B An example is shown of the distance-velocity characteristic curve and the tolerance zone associated with the reasonableness value. Detailed Implementation

[0076] Subsequently, Figure 1 The schematic diagram of system 1 refers to embodiments 3a, 3b, and 3c of the method for identifying following vehicle situations. Figures 2A to 2C Steps 31, 32a, 32b, and 32c shown in the figure are described.

[0077] System 1 includes a distance sensor system 12 for detecting the actual following distance between the vehicle and a vehicle traveling in front (the vehicle ahead). For example, the distance sensor system 12 may have one or more radar sensors. In addition to following distance, the radar sensors can also be used to determine the relative speed between the vehicle and the vehicle ahead.

[0078] In addition, system 1 also includes a speed sensor system 13 for detecting the actual speed of the vehicle. The speed sensor system 13 may include, for example, one or more wheel speed sensors, one or more acceleration sensors (from which the actual speed can be calculated) and / or a receiver for a global navigation satellite system (wherein the actual speed can be determined based on changes in vehicle position detected by means of the global navigation satellite system).

[0079] The (data) processing device 10 of System 1 is connected to the distance sensor system 12 and the speed sensor system 13 in signal technology, and is designed to receive a distance signal A, including the actual following distance, from the distance sensor system 12, and a speed signal G, including the actual speed, from the speed sensor system 13. The processing device 10 also receives a relative speed signal V from the distance sensor system 12, which can be combined with the speed signal G to determine the speed of the vehicle ahead.

[0080] Furthermore, the processing device 10 is connected to the data memory 11 of the system 1 in terms of signal technology and is designed to receive characteristic curve information K (e.g., speed value and target following distance value associated with the speed value) from the data memory based on the distance-speed characteristic curve (hereinafter also referred to as: characteristic curve) stored therein.

[0081] The processing device 10 is also designed to determine updated characteristic curve information K' based on the detected driving behavior (especially based on the detected actual distance and actual speed) and output it to the data memory 11 for storage, so that the updated characteristic curve is stored in the data memory 11.

[0082] The data storage 11 is also connected to the control device 14 in signal technology, which is designed to control the automatic longitudinal guidance of the vehicle. The control device 14 can, for example, read characteristic curve information K' from the data storage 11 according to the updated characteristic curve, and generate a control signal S for controlling the automatic longitudinal guidance according to the updated characteristic curve information K', in particular to adjust the target following distance to the vehicle ahead according to the updated characteristic curve information K'.

[0083] exist Figure 1 In the illustrated embodiment, the processing device 10, designed to identify following conditions, is separate from the control device 14 that controls the vehicle's automatic longitudinal guidance. Here, the control device 14 is connected to the processing device 10 in signal technology and receives distance and speed information A, G, V from it as the basis for the vehicle's longitudinal adjustment.

[0084] However, the following implementation is also possible, wherein a single processing device identifies the following situation and subsequently adjusts automatic longitudinal guidance based on a characteristic curve learned during the following situation and / or based on the actual following distance detected during the following situation. In other words, refer to Figure 1 The processing device 10 may be the same as or a part of the control device 14 (or vice versa).

[0085] The control device 14 can output a control signal S to the vehicle's longitudinal guidance actuator 15, particularly the braking equipment and / or drive module. The longitudinal guidance actuator 15 can control automatic longitudinal guidance according to the control signal S, and in particular adjust the target following distance to the vehicle ahead. Figure 1 In the diagram, the longitudinal guide actuator 15 is drawn with a dashed line because it can be a general longitudinal guide actuator known in the prior art, and need not be understood as part of the system 1 according to the invention.

[0086] The processing device 10 of System 1 can be designed to adapt the distance-speed characteristic curve based on the actual following distance, within the range of a learning algorithm for learning the following distance for automatic longitudinal guidance of the vehicle during the learning phase.

[0087] Figure 3A An example of this (distance-velocity) characteristic curve is shown graphically, in which the second interval d is plotted against the velocity v. Here, the actual characteristic curve is shown in the graph as small circles connected by solid lines (the "base" in the graph's description). The data points shown as small circles assign the second interval to the support point velocity (hereinafter also referred to as the support point). Between the support points, the characteristic curve is generated by (linearly) interpolation.

[0088] exist Figure 3A The exemplary characteristic curve shown uses five support points. The more support points there are, the more personalized the characteristic curve becomes. Therefore, if available storage space allows for this in a specific application, it is theoretically meaningful to use significantly more than five support points. However, based on the interpolation that is eventually performed between the various data points, it is not necessary to choose a step size of, for example, 1 km / h or smaller between the support points in order to obtain important results.

[0089] The most common speed limits, such as 30, 50, 80, 100, and 120 km / h, and some higher speeds, such as 140, 160, or 180 km / h, are sufficient to provide a perceptible effect for the driver. If we add the speed limits of the system (i.e., especially the autonomous driving function) to this, then for example, we get 10 support points to be stored and 10 related distance values ​​to be stored.

[0090] However, according to an advantageous implementation, while individual second intervals cannot be learned entirely freely, certain boundaries can be set, and the movement must occur within these boundaries. Figure 3A In the diagram, this is illustrated exemplarily using lower and upper boundary characteristic curves (see the description: "Boundary"). The characteristic curve being learned must lie between the upper and lower boundary characteristic curves.

[0091] For example, the upper limit can be provided by the performance of the sensor system. Conventional radar sensors have an effective range of approximately 150 meters, while more advanced components have effective ranges of up to 300 meters. In the case of objects at greater distances, target loss may occur, potentially leading to unpleasant longitudinal adjustment behavior perceived by the driver. If distant targets are repeatedly lost, it results in acceleration and braking that the driver cannot understand. In following a vehicle with a target traveling at 190 km / h, the following distance corresponding to the radar's maximum effective range of 150 meters corresponds, for example, to a distance of 2.84 seconds. However, to avoid erroneous braking or other undesirable effects, it is preferable to also incorporate additional buffers. Therefore, according to... Figure 3A In this embodiment, a value of 2.5 seconds is assumed as the maximum interval, which forms the upper boundary characteristic curve. Depending on the sensor performance, the upper boundary characteristic curve can be further increased or decreased.

[0092] For the lower boundary, both legal regulations and standard conditions can be considered as boundary conditions. For example, the distance derived from the learned characteristic curve should not be lower than the distance derived from the characteristic curve corresponding to the minimum distance level set by the driver. If the ACC function has, for example, three distance levels, then the minimum distance level should form the lower boundary. Figure 3A The lower boundary characteristic curve in the curve corresponds to the defined boundary.

[0093] In a system with three selectable distance levels, the characteristic curve stored for the intermediate distance level can, for example, be used as a starting point for learning the characteristic curve. This characteristic curve can serve as a known reference point for the driver, who may have also driven previous models of the same vehicle brand. Based on this, learning (regarding following distance) in both positive and negative directions can be quickly achieved, allowing the driver to perceive the positive impact in a timely manner.

[0094] As mentioned above, specific driving conditions or time segments can be identified as suitable for the learning characteristic curve during driving. This can especially apply to following situations where the following distance is relatively constant and is controlled by the driver.

[0095] Figures 2A-2CEach of the following is illustrated in the form of a schematic block diagram, showing the steps 31, 32a, 32b, and 32c of the corresponding implementation variants 3a, 3b, and 3c of the method for identifying following situations.

[0096] In step 31 of the common method for all implementation variants 3a, 3b, and 3c, it is determined whether a following situation exists based on one or more predetermined criteria. In a following situation, the vehicle follows the target object (especially the vehicle in front) at a following distance that may be anticipated by the driver, under the influence of the driver's operating behavior.

[0097] The processing device 10 can, for example, execute method step 31 within the scope of a following vehicle recognition algorithm, particularly based on signals A, G, and V transmitted to it. Here, a series of criteria can be used to determine whether a following vehicle situation exists.

[0098] For example, distance signal A can be used as a prerequisite for determining whether there is a following situation in vehicle 31, and the actual following distance detected can be checked to see if it is within a predetermined range.

[0099] Alternatively or additionally, as a condition for determining whether following is occurring, the acceleration of the vehicle and / or the target object can be checked to see if they are within a predetermined range. To determine the acceleration of the vehicle or the target object, the velocity signal G or the relative velocity signal V (from which the velocity signal G can be combined to determine the velocity of the target object) can, for example, be differentiated over time.

[0100] Another possible criterion could directly relate to the difference between the target object's speed and the vehicle's speed, provided by the relative speed signal V. For example, it could be used as a prerequisite for determining whether following another vehicle exists (e.g., checking if this difference is less than (or less than or equal to) the predetermined maximum speed difference).

[0101] Determining whether vehicle 31 is following another vehicle can also be done by identifying any unusual driving situations.

[0102] Special driving situations in this sense may involve a vehicle approaching a traffic light or intersection, or (e.g., identified due to manipulation of a direction indicator or active navigation route guidance) an impending vehicle turn. In such cases, it can be assumed that the driver primarily guides the vehicle longitudinally based on the aforementioned events, such that the resulting actual following distance cannot be assumed to represent the expected distance normally observed in following situations. Accordingly, it can be stipulated that following situations will not be determined in the aforementioned special driving situations.

[0103] In addition to the above standard, as a condition for determining that 31 is in a following situation, it can be checked whether the actual following distance detected for the vehicle speed deviates from the following distance set for that speed according to the stored distance-speed characteristic curve by no more than a predetermined size.

[0104] As in Figure 3B As exemplarily and schematically illustrated, the tolerance band T can, for example, be placed around a stored characteristic curve (which may have been learned over a certain period of time). The tolerance band T can have different widths for different speeds, wherein the width of the tolerance band does not necessarily have to be symmetrical (with respect to deviations upward and downward along the distance axis). Figure 3B The tolerance band T width B is plotted for illustrative purposes. This width represents the maximum permissible positive deviation of the actual following distance at a speed of 50 m / s compared to previously stored characteristic curves. If the actual following distance detected at 50 m / s is large enough that it exceeds the second interval set according to the previously stored characteristic curves by a predetermined size (i.e., width B), then following distance is not determined.

[0105] Tolerance zone T or more generally, predetermined dimensions (e.g.) Figure 3B The width B in the tolerance band and the corresponding width of the tolerance band T for downward distance deviation or upward and / or downward distance deviation for other speeds can be correlated with the consistency of driving behavior already stored in the stored distance-speed characteristic curve.

[0106] The longer a driver uses the vehicle on the road, the more information is known about the driver's intentions. Based on this, the reasonableness or reliability of the values ​​learned so far can be determined. If the current actual following distance deviates too far from the learned values, it can be deduced that there is no true following situation, and this can, for example, prevent the characteristic curve from learning (further) based on the current actual following distance.

[0107] By employing this reasonableness check, obvious or accidental following can be identified, especially when considering the validity of the characteristic curves learned to date, and therefore not treated as actual following situations. This could, for example, involve a scenario where a driver is traveling at 50 km / h in a city according to a speed limit. Another vehicle, also adhering to the speed limit, is traveling at a greater distance. Neither deviates significantly and therefore chooses the same desired speed. However, the driver of the learning vehicle is not intentionally following the vehicle ahead at a specific distance; instead, the first priority here is to adhere to the speed limit. Another example scenario is following a tractor that has lost a portion of its cargo. The driver does not want to come into contact with the cargo and therefore maintains a significantly greater distance. Because in both cases, the two vehicles are constantly traveling in sequence, acceleration is minimal, and the speeds of the vehicles do not deviate from each other, these situations can be primarily represented as relevant following situations based on some of the aforementioned criteria. However, in both of these example scenarios, the currently detected actual following distance deviates unreliably from the driver's desired distance by an excessive amount, as would be detected in other situations. Based on this deviation, these situations can be identified as non-following situations.

[0108] For example, a predetermined size (e.g.) Figure 3B The width B of the tolerance zone T can be set according to a reasonableness value k, which is increased or decreased by comparing the difference between the actual following distance detected at the corresponding time point and the target following distance applicable according to the distance-speed characteristic curve at the corresponding time point at a previous multiple time points with the last observation time point. Specifically, it can be stipulated that the larger the reasonableness value k, the smaller the predetermined size is set (and vice versa). This can be achieved, for example, by having the width B of the tolerance zone determined by a mathematical expression in which at least one term is inversely proportional to the reasonableness value k or the reasonableness value k is subtracted from a constant such that the term decreases linearly as the reasonableness value k increases.

[0109] A reasonableness value k can be defined, for example, as a value between 0 and 1 with a fixed resolution, where 0 indicates no learning and 1 represents maximum reasonableness. The reasonableness value k should depend not only on the learning time but also on how well the actual distance during the learning period corresponds to the time interval that has been learned. If the current distance is very similar to the time interval that has been learned, then the reasonableness value k can increase; if the new current distance deviates significantly from the time interval that has been learned so far, then the reasonableness value can decrease. Therefore, the reasonableness value k represents not only the learning time but can also be interpreted as a statement of the reliability of the currently stored characteristic curve.

[0110] The increase or decrease of the reasonableness value k can be based on the actual following distance d being detected.ist The characteristic curve learned up to that point in time is applicable to the speed v of travel. ist Target following distance d gelernt The reasonableness value k is determined by the deviation between the two values. If the difference is small during the active learning phase, the reasonableness value k increases more. If the difference is large, the reasonableness value k increases or even decreases by a smaller value. If d gelernt Approaching d due to the learning algorithm ist Then k increases again.

[0111] Additionally determine d gelernt The surrounding range of values ​​within which the rationality k either increases or decreases. This can be achieved by presetting a fixed distance value d. a The following is performed, where the actual following distance d ist Compared to the following distance d that has been learned so far gelernt Deviation greater than distance value d a Therefore, the rationality k should be reduced.

[0112] The calculation of the updated rationality value k' based on the previous (e.g., determined in a previous calculation cycle) rationality value k can be implemented, for example, using the following algorithm:

[0113] Here, the latter part describes the distance d that depends on the distance being learned. gelernt The actual distance d at present ist A function representing the difference between the given distance values. This function is exactly at the set distance value d. a It has its zero point. By choosing the factor α d We can presuppose how quickly the value k should actually change. As mentioned above, the reasonable value k can, for example, be limited to a range between 0 and 1 to always allow for further learning.

[0114] Preferably, the rationality value k is continuously stored throughout multiple learning stages (and, if necessary, during the driver's multiple drives), and continues to increase or decrease accordingly.

[0115] However, it's important to note that a driver's distance preference is not the same on every trip. Emotions and stress can influence the current perception of distance. What might have been the highest level of rationality (k) on the last trip may not be high on the next.

[0116] Therefore, according to the implementation method, the stored reasonableness value k is limited. This should be illustrated in the example scenario: during the final drive, the driver moves within the same value range, around a following distance of 2 seconds, so that at the end of the drive, the reasonableness value k is close to 1, which excludes large distance deviations. If the driver now wants to travel a smaller distance of approximately 1.3 seconds in a new drive, then he will be outside the limited range, and thus it cannot be learned. This problem can be solved by reducing the reasonableness value k to, for example, 0.7 after each drive (or at the start of a new drive). In the example scenario, the current second distance is again within the range where learning is possible. This solution provides the driver with greater flexibility at the start of a drive and thus allows for responses to changes in distance expectations. However, the limitation or reset of the reasonableness value k after the drive ends (or at the start of a new drive) should be determined such that obviously erroneous scenarios are still not learned.

[0117] In general, the reasonableness value k can be used to state how reliable the currently learned distance is. Based on this, for example, it is possible to suppress the learning of erroneous distances that are far from the distances learned so far.

[0118] The rationality value k can be calculated for the entire characteristic curve. However, it is also possible to perform the calculation individually for each or all support points to make the distinction more accurate.

[0119] If the change in the target object is determined, it can be used as a condition to determine whether there is a following situation in 31. Check whether the predetermined minimum duration has passed since the change in the target object was determined.

[0120] A change in the target object can occur, for example, when another vehicle inserts itself between the current vehicle and the previous vehicle (the previous target object), thus becoming the new target object. This corresponds to the experience that the current vehicle's driver needs a certain amount of time, depending on the situation, to adjust its desired distance from the new target object. Therefore, it may be meaningful to wait for a minimum duration before re-establishing following distance. This minimum duration could, for example, be in the range of 2 to 12 seconds.

[0121] The minimum duration can be dynamically set based on parameters representing the situation, such as the speed of the vehicle and / or the new target object and / or whether the vehicle is in an environment within or outside the city. Therefore, the more dynamic the situation, the longer the time lag (i.e., the minimum duration) after the target object changes.

[0122] System 1 includes an output device 16 and is designed to output instructions to the driver via the output device 16 when the processing device 10 determines that a following situation exists, so as to suggest to the driver the use of an automated driving function (such as ACC) with automatic longitudinal guidance.

[0123] At the method level, this corresponds to according to Figure 2A In implementation variant 3a, the processing device 10 generates a control signal H for the output device 16 in step 32a to trigger a corresponding indication. Figure 1 In this embodiment, output device 16 is exemplarily and schematically shown as a speaker, since the indication can be output in an acoustic form (e.g., as voice output). However, alternatively or additionally, the output device may also include, for example, devices for optical and / or tactile indication output, such as a display, a steering wheel vibration device, a seat vibration device, etc.

[0124] exist Figure 2B In the implementation variant 2b of the method shown, after determining the following situation 31, in a further step 32b, the actual following distance detected in the following situation is used for automatic longitudinal guidance of the vehicle within the range of the learning algorithm used to learn the following distance. After determining that the following situation 31 exists, for example, the learning phase of the learning algorithm can be initiated. The learning phase can continue as long as (e.g., in each additional calculation cycle) it is determined that the following situation still exists.

[0125] When the learning phase of the learning algorithm used to learn the following distance is active, the driver can use output device 16 or other means (in...) Figure 1 The output device (not shown separately) is informed of this situation. Here, the driver may be additionally informed, for example, of the current "direction" in which the learned following distance is being adapted, i.e., whether the driver's current driving behavior will cause an increase or decrease in the previously stored target following distance.

[0126] Learning algorithms may include, in particular, adapting to the actual following distance detected. Figures 3A-3B The distance-speed characteristic curve of the type shown is described above. As mentioned above, to adapt to the distance-speed characteristic curve, for example, a method can be implemented, which is the subject of a patent application filed by the applicant with the German Patent and Trademark Office on January 15, 2024, entitled "Method for Learning Following Distance and System for Automatic Longitudinal Guiding of Vehicles".

[0127] As described above, the processing device 10 can receive characteristic curve information K from the data memory 11 when executing the learning algorithm, and determine updated characteristic curve information K' based on the actual following distance detected in the following driving situation. The updated characteristic curve information K' can be stored in the data memory 11 and recalled by the control device 14. Subsequently, the control device 14 can generate a control signal S for the longitudinal guidance actuator 15 based on the updated characteristic curve information K', so as to adjust the target following distance with the vehicle in front according to the updated characteristic curve information K'.

[0128] Typically, in a manner different from adapting the characteristic curve within the range of the learning algorithm described above, System 1 can be designed to perform automatic longitudinal guidance of the vehicle in such a way that the following distance to the target object is set based on the actual following distance detected in the identified following driving situation.

[0129] At the method level, this aspect corresponds to... Figure 2C In the implementation variant 3c, the control device 14 generates a control signal S for the longitudinal guide actuator 15 based on the distance signal A in step 32c.

[0130] When an autonomous driving function with automatic longitudinal guidance of the vehicle is activated after a following situation is determined, the control device 14 can set a following distance to the target vehicle based on the actual following distance detected during the following situation as at least temporarily deviating from the target following distance, which is typically applicable to the autonomous driving function at the relevant speed according to a stored distance-speed characteristic curve. Within the scope of the autonomous driving function, the following distance can be set, for example, in a way that gradually approaches the target following distance. When the driver, for example, first deliberately drives with a relatively large following distance from the vehicle in front without ACC activated and then activates ACC, the control device 14 can adapt the longitudinal adjustment in the sense of driver comfort, so that the following distance traversed by ACC gradually approaches the target following distance.

Claims

1. A computer-implemented method (3a, 3b, 3c) for identifying following vehicle situations, comprising the following steps: - Determine (31) whether a following situation exists by means of one or more predetermined criteria, in which the vehicle follows the target object at a following distance expected by the driver under the influence of the driver's operation behavior of the vehicle; - When it is determined that (31) there is a situation of following another vehicle, ○ Trigger (32a) an instruction to the driver to suggest an automated driving function to the driver; and / or ○ The actual following distance detected in the aforementioned following driving situation ■ Within the scope of the learning algorithm used to learn (32b) following distance, automatic longitudinal guidance for the vehicle, and / or ■ Used to set (32c) following distance within the range of automatic longitudinal guidance of the vehicle.

2. The method according to claim 1 (3a, 3b, 3c), wherein, As a condition for determining (31) the existence of following traffic, it is checked whether the actual following distance detected is within a predetermined value range.

3. The method according to any one of the preceding claims (3a, 3b, 3c), wherein, As a condition for determining (31) the existence of following vehicle driving, it is checked whether the acceleration of the vehicle and / or the acceleration of the target object are within a predetermined value range.

4. The method according to any one of the preceding claims (3a, 3b, 3c), wherein, As a condition for determining (31) the existence of following traffic, it is checked whether the absolute value of the difference between the speed of the target object and the speed of the vehicle is less than a predetermined maximum speed difference, and / or wherein, As a condition for determining (31) the existence of following vehicle situation, it is checked whether the difference between the speed of the target object and the speed of the vehicle is less than or equal to the predetermined maximum speed difference in absolute value.

5. The method according to any one of the preceding claims (3a, 3b, 3c), wherein, Determine the change of the target object, and wherein, as a prerequisite for determining (31) the existence of following vehicle driving, check whether a predetermined minimum duration has elapsed since the change of the target object was determined.

6. The method according to any one of the preceding claims (3a, 3b, 3c), wherein, Determining (31) whether there is a following vehicle situation is based on whether special driving situations are identified, wherein the special driving situations include the following: - The vehicle proceeded toward the traffic light; and / or - The vehicle is heading toward the intersection; and / or - The vehicle is about to make a turn.

7. The method according to any one of the preceding claims (3a, 3b, 3c), wherein, As a condition for determining (31) the existence of following traffic, it is checked whether the actual following distance detected for the speed of the vehicle deviates from the following distance set for the speed according to the stored distance-speed characteristic curve by no more than a predetermined size.

8. The method according to claim 7 (3a, 3b, 3c), wherein, The predetermined size is set according to a reasonableness value, wherein the reasonableness value is increased or decreased at multiple previous time points based on the difference between the actual following distance detected at the corresponding time point and the target following distance applicable according to the distance-speed characteristic curve at the corresponding time point, compared with the last observed time point.

9. A processing apparatus (10) having one or more processors, wherein, The processing device (10) is designed to perform the method (3a, 3b, 3c) according to any one of the preceding claims by means of the one processor or by means of the plurality of processors.

10. A system for identifying following vehicle situations (1), wherein, The system (1) includes a processing device (10) according to claim 9, or is connected or connectable to such a processing device (10) via data technology, and wherein the system (1) is designed to, when it is determined (31) that a following vehicle situation exists, - Output instructions to the driver via output device (16) to suggest autonomous driving functions to the driver; and / or The automatic longitudinal guidance of the vehicle is performed in such a way that the following distance to the target object is set based on the actual following distance detected during the following driving situation.

11. The system according to claim 10, wherein, The system (1) is designed to set a following distance from the target object based on the actual following distance detected in the following situation, which is a deviation from the target following distance, if an automatic driving function with automatic longitudinal guidance of the vehicle is activated after the following situation is determined, the following distance is applied to the automatic driving function according to a stored distance-speed characteristic curve.

12. The system (1) according to claim 11, wherein, The system (1) is designed to set the following distance in such a way that the following distance approaches the target following distance over time within the scope of the autonomous driving function.

13. A vehicle having the system (1) according to claim 12.

14. A computer program comprising commands that, when executed by a processing device (10), cause the processing device to perform the method (3a, 3b, 3c) according to any one of claims 1 to 8.

15. A computer-readable storage medium comprising a command that, when executed by a processing device (10), causes the processing device to perform the method (3a, 3b, 3c) according to any one of claims 1 to 8.

Images