System and method for learning a following distance for automated longitudinal guidance of a vehicle

Abstract

A system (1) for learning a following distance for automated longitudinal guidance of a vehicle is configured to adapt a normal operation following distance (N) to a target object traveling ahead of the vehicle, provided for normal operation of the automated longitudinal guidance, as a function of a detected override of the automated longitudinal guidance by a driver of the vehicle, so that a modified normal operation following distance (N) is provided for future following travel during normal operation of the automated longitudinal guidance. The system is further configured to set a learning operation following distance (L) to a target object traveling ahead, which is increased compared to the normal operation following distance (N), depending on the situation, so that the driver has more opportunity to set a desired following distance that is reduced compared to the learning operation following distance (L) by overriding the automated longitudinal guidance.

Description

[0001] The invention relates to a system for learning a following distance for automated longitudinal guidance of a vehicle and to a vehicle equipped with such a system. Furthermore, the invention relates to a computer-implemented method for learning a following distance for automated longitudinal guidance of a vehicle, a processing device and a computer program for executing such a method, and a computer-readable storage medium on which such a computer program is stored.

[0002] Many modern motor vehicles are equipped with automated driving functions which can in particular enable automated longitudinal guidance of the vehicle. One example of a widely used automated driving function is automatic distance control, which is often also referred to as adaptive cruise control or ACC (for “Adaptive Cruise Control”). If a vehicle in front is traveling slower than a target speed set for the vehicle, the following distance to the vehicle in front is regulated to a target following distance using automatic engine and braking interventions. The target following distance can, for example, be a specific second interval, i.e. it can be specified as a specific time interval at which the vehicle equipped with ACC should follow the vehicle in front.

[0003] Since different drivers prefer different following distances, the target following distance is usually adjustable by the driver within certain limits. For example, the driver may be able to set the following distance globally in a number of distance levels, with a respective distance-speed characteristic curve stored for each of the distance levels. However, such predefined distance levels only partially reflect individual needs and habits. While they reflect an average value for average drivers, they do not offer the option of accommodating specific driver preferences.

[0004] The course of the preset distance-speed characteristic curves follows a fixed pattern in conventional systems. At low speeds, the interval per second is usually higher. The higher the speed, the smaller the corresponding distance becomes. However, this does not necessarily correspond to the driver's desired behavior. For example, a driver may like to follow closely in heavy city traffic and choose comparatively large intervals on motorways and country roads in order to reach their destination in a relaxed manner. These preferences are not reflected in conventional distance levels, as this does not correspond to the average driver. Due to the limited adjustability to a few distance levels, individual driver preferences can only be taken into account to a limited extent.

[0005] There are already known solutions in the state of the art to learn desired distances from a driver's driving behavior and to take these into account in automatic longitudinal guidance.

[0006] For example, it has been proposed to learn individual driver behavior with regard to following distances using neural networks (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, 95th vol., pp. 346-362.).

[0007] DE 10 2015 016 993 A1 describes a method for determining a driving profile for a driver assistance system of a vehicle, comprising the steps of: when the driver assistance system is inactive, recording distance measurement signals from a distance measurement system of the vehicle and driving dynamics measurement signals of the vehicle; storing the recorded distance measurement signals and driving dynamics measurement signals in a memory device; determining a driving behavior from the stored distance measurement signals and driving dynamics measurement signals, wherein a distance and / or temporal distance to a vehicle in front is determined from the distance measurement signals; and storing the determined driving behavior. Furthermore, a driver assistance control method, in particular an ACC control method using the method, and a driver assistance system for implementing the driver assistance control method are proposed.

[0008] Conventional approaches, such as the one mentioned in the previous paragraph, only allow for learning the driver's manual driving style with adaptive cruise control deactivated. However, in Germany, for example, over 30% of the kilometers driven in ACC-equipped vehicles are already covered with adaptive cruise control activated, meaning that such conventional solutions do not allow for adjustments to the driver's distance preferences on these significant portions of the route.

[0009] However, especially among users who frequently use the ACC system and are therefore most likely to evaluate and question the automatic learning of distance behavior, there is a need to effectively implement desired adjustments to distance behavior. Therefore, it is desirable to adapt to the driver's instructions even during phases when distance control is activated.

[0010] A practical example is a driver who likes to use the ACC system on country roads and highways, but prefers manual control of the vehicle in urban scenarios due to the highly dynamic driving situations. If the ACC system uses different distance-speed characteristics for city and highway scenarios, the urban characteristic will be sufficiently learned due to the high proportion of manual driving. However, if the driver activates the adaptive cruise control as soon as they enter a highway, a standard characteristic for the highway or a characteristic that is only minimally adapted from this standard characteristic due to the low proportion of manual driving on the highway would always be used.

[0011] Therefore, there is a need for a technical solution that enables adaptation to the driver's habits even during active driver assistance, especially on a motorway or motorway-like road.

[0012] It is an object of the present invention to provide a computer-implemented method for effectively learning a following distance based on driver behavior detected when the distance control is activated, as well as a corresponding system that can carry out such a method.

[0013] The problem is solved by the subject matter of the independent patent claims. Advantageous embodiments are specified in the dependent patent claims.

[0014] A first aspect of the invention relates to a system for learning a following distance for automated longitudinal guidance of a vehicle.

[0015] The vehicle can, in particular, be a motor vehicle. The term "motor vehicle" is understood to mean, in particular, a land vehicle that is propelled by mechanical power without being tied to railway tracks. A motor vehicle in this sense can be, for example, a passenger car, motorcycle, or tractor.

[0016] The automated longitudinal guidance of the vehicle can be carried out as part of an automated driving function, such as an ACC function or a more comprehensive function that supports, for example, combined longitudinal and lateral guidance.

[0017] In general, the term "automated driving function" in this document refers to a vehicle function that enables automated driving. The term "automated driving" refers to driving with automated longitudinal and / or lateral guidance. This can, for example, involve extended driving on the highway or limited-time driving when parking. The term "automated driving" encompasses automated driving with any degree of automation. Examples of automation levels include assisted, partially automated, conditionally automated, highly automated, and fully automated driving (with increasing levels of automation in each case). The five levels of automation mentioned above correspond to SAE Levels 1 to 5 of the SAE J3016 standard (SAE - Society of Automotive Engineering) as of April 30, 2021.In assisted driving (SAE Level 1), the system performs longitudinal or lateral guidance in certain driving situations, with the expectation that the driver will assume all remaining aspects of the dynamic driving tasks. In partially automated driving (SAE Level 2), the system assumes longitudinal and lateral guidance in certain driving situations, whereby the driver must continuously monitor the system, as in assisted driving. In conditionally automated driving (SAE Level 3), the system assumes longitudinal and lateral guidance in certain driving situations without the driver having to continuously monitor the system; however, the driver must be able to assume control of the vehicle within a certain period of time upon request from the system. In highly automated driving (SAE Level 4), the system assumes control of the vehicle in certain driving situations, even if the driver does not respond to a request for intervention, thus eliminating the driver as a fallback.In fully automated driving (SAE Level 5), the system can perform all aspects of the dynamic driving task under any road and environmental conditions that can also be mastered by a human driver.

[0018] The system according to the invention is configured to adapt a normal operating following distance to a target object driving ahead of the vehicle (in particular to a vehicle in front) provided for normal operation of the automated longitudinal guidance as a function of an oversteering of the automated longitudinal guidance by a driver of the vehicle detected in a learning phase of the automated longitudinal guidance, so that a modified normal operating following distance is provided for a future following journey in normal operation of the automated longitudinal guidance.

[0019] The system is therefore capable of learning with regard to the normal operating following distance, for example, by storing a new normal operating following distance for use during a subsequent following journey in normal operation. The new normal operating following distance is modified from the previous normal operating following distance based on detected oversteering behavior by the driver. The previous normal operating following distance can itself already be the (intermediate) result of such a learning process. Alternatively, the previous normal operating following distance can be, for example, a preset standard following distance.

[0020] For example, the normal operating following distance can be speed-specific. In particular, the normal operating following distance can result from a preset or learned distance-speed characteristic curve (hereinafter also referred to as "characteristic curve") and, for example, be assigned to a current speed of the vehicle as the target following distance according to this characteristic curve. Learning the driver's distance behavior does not have to be limited to a single normal operating following distance, but can occur for such a characteristic curve as a whole. Details on learning a distance-speed characteristic curve are explained further below in connection with possible embodiments of the system according to the invention and a method according to the invention.

[0021] Overriding the automated longitudinal guidance can, in particular, involve shortening the following distance by the driver actuating the accelerator pedal. Using the accelerator pedal, the driver can reduce the distance to the vehicle in front as desired, even when the distance control is activated. This is referred to as oversteering, in which the longitudinal dynamics requirement of the distance control is overridden by the driver. The changed (new) normal operating following distance can then, in particular, be a shorter distance compared to the previous normal operating following distance. The changed normal operating following distance can thus take into account the driver preference for a shorter following distance in a future following journey (and possibly even during the further course of the current following journey), expressed by the driver actively shortening the following distance.In this way, the distance behavior of the automated longitudinal guidance system desired by the driver can be gradually learned better and better.

[0022] The system is further configured to set a learning operation following distance to a target object ahead that is larger than the normal operation following distance, depending on the situation. The learning operation following distance is so named because it is intended to enable effective adaptation of the normal operation following distance to the driver's preferences due to more detected driver oversteering interventions.

[0023] The invention is based on the idea that active adjustment of the following distance by the driver through overriding the accelerator pedal can only result in a reduction of the stored normal operating following distance. Without the solution according to the invention, drivers with a greater desired distance (compared to the previous normal operating following distance) must therefore generally deactivate the ACC system by braking in order to increase the following distance. Therefore, only drivers with a lower desired distance can actively provide impulses for learning the following behavior.

[0024] By increasing the initial following distance to the learning mode following distance proposed here, a larger number of drivers can set their desired distance and thus train the automated longitudinal guidance to their preferred distance behavior. In particular, the system can deliberately set a larger learning mode following distance than the normal mode following distance intended for normal operation, at least initially—i.e., immediately after activation of the distance control or the occurrence of a following situation—in order to subsequently detect whether and, if so, how much the driver reduces the following distance by pressing the accelerator pedal.

[0025] The increased learning mode following distance enables effective learning of the distance behavior of the automated longitudinal guidance system even when the adaptive cruise control is activated, because the driver has more opportunity to set a desired following distance that is shorter than the learning mode following distance by overriding the automated longitudinal guidance. This means that (at least statistically averaged across all driver preferences) there will be a more frequent desire to shorten the following distance with the increased learning mode following distance than with the normal mode following distance. As a result, more override interventions are available as a data basis for learning the normal mode following distance.

[0026] According to one embodiment, the statement that the learning mode following distance is set "situationally" can be understood in particular to mean that the system must first determine that at least one switching condition is met as a prerequisite for setting the learning mode following distance instead of the smaller normal mode following distance. The switching condition is named this way because it can be understood as a condition for switching between the normal mode following distance and the learning mode following distance as the following distance actually to be set. As will be described further below with reference to an embodiment variant of the method proposed here, such a switching condition can, for example, relate to a road type currently being traveled on. For example, a switching condition can be that the vehicle is currently traveling on a motorway or a motorway-like road.

[0027] According to one embodiment, the system is configured to set the learning mode following distance only temporarily and to gradually shorten the following distance to the target object ahead, starting from the learning mode following distance, toward the normal mode following distance, as long as the driver does not override the automated longitudinal guidance. This can prevent a driver who does not want to actively override the automated longitudinal guidance from being forced by the system to drive for a longer period at a following distance that is likely too large for their perceived distance.

[0028] A second aspect of the invention is a vehicle, in particular a motor vehicle, with a system according to the first aspect of the invention.

[0029] A third aspect of the invention is a computer-implemented method for learning a following distance for automated longitudinal guidance of a vehicle.

[0030] The method can be carried out, for example, by means of a system according to the first aspect of the invention, in particular by means of one or more (data) processing devices of such a system. In this respect, the above and following explanations regarding the system according to the invention and its possible configurations apply analogously to possible embodiments of the method according to the invention, and vice versa.

[0031] The vehicle whose automated longitudinal guidance is to be adjusted by the method comprises a processing device which is configured to set a normal operating following distance to a target object driving ahead during normal operation of the automated longitudinal guidance, i.e., for example, to generate corresponding control signals for the automated longitudinal guidance.

[0032] The method comprises the steps explained below, which can be carried out by means of the above-mentioned processing device and / or by means of one or more further processing devices.

[0033] In one step, a learning operation following distance to a target object ahead is set which is increased compared to the normal operation following distance.

[0034] In a further step, any override of the automated longitudinal guidance by a vehicle driver, which reduces the following distance compared to the learned mode following distance, is recorded. Particularly relevant are those scenes in which the driver actively overrides the automated longitudinal guidance using the accelerator pedal in order to reduce the following distance. If the driver requests greater acceleration than the automated longitudinal guidance system in this way, the driver controls the vehicle's longitudinal guidance in a similar way to how they would without activated distance control, so that the set following distance can be considered informative for the preferred following behavior, just like in purely manual driving (possibly under other conditions that may, for example, relate to the recognition of an actual following scenario).

[0035] In a further step, the normal operating following distance for future following journeys during normal operation of the automated longitudinal guidance system is adjusted depending on the detected oversteer of the automated longitudinal guidance system. For example, the normal operating following distance can be stored as part of a distance-speed characteristic curve, which is successively adjusted to the driver's desired behavior based on detected oversteer events.

[0036] According to a further development, the method can further comprise, as an additional step preceding the aforementioned steps, checking whether a switching condition is met. In this case, it can be provided that the learning operation following interval is only set (instead of the normal operation following interval) if the check shows that the switching condition is met. The switching condition can also be a combined switching condition composed of several individual conditions. For example, it can be provided that the switching condition is only considered met if the several individual conditions are met cumulatively.

[0037] For example, the switching condition (e.g., as one of several individual conditions) may include the vehicle traveling on a specific road type, in particular a motorway or a motorway-like road. This circumstance may be determined, for example, in a conventional manner based on map information from a navigation system and / or sign recognition.

[0038] As already mentioned, the invention can be profitably applied particularly on motorways or motorway-like roads. There, due to the high proportion of phases with active distance control, the number of scenes in which the driver sets their desired distance without active distance control may be too small to effectively adapt the normal operating distance to the driver's distance preferences on this basis alone. The method according to the invention can remedy this situation by providing for learning even with active driver assistance, thereby giving the driver more options for setting their desired following distance by setting the increased learning operating distance by overriding the distance control using the accelerator pedal.

[0039] Alternatively or in addition to being dependent on the road type currently being traveled, the switching condition can refer to certain measured variables relevant to a following journey. For example, the switching condition can include an actual following distance being within a predefined range and / or an actual speed of the vehicle, an actual speed of the target object, and / or a relative speed between the target object and the vehicle in front being within a respective predefined range.

[0040] According to one embodiment, adapting the normal operating following distance as a function of the detected oversteering comprises executing several steps for at least one interpolation point speed of a distance-speed characteristic curve during a learning phase, which steps are explained in detail below. According to one embodiment variant, the steps can each be executed for several interpolation point speeds of the distance-speed characteristic curve. Alternatively or additionally, it can be provided that the steps for the interpolation point speed or for the several interpolation point speeds are each executed at several times during the learning phase.

[0041] The basic approach used to adapt the adaptive cruise control to individual driving behavior involves learning specific distance characteristics. The goal is to memorize the driver's preferences and adjust the set distance to their needs. The adaptive cruise control system is thus adapted to the driver's personal driving profile to make it as comfortable as possible for the driver.

[0042] Specifically, this is to be achieved by automatically adapting a stored distance-speed characteristic curve, which is used within the framework of automated longitudinal guidance for speed-dependent adjustment of the distance to the vehicle in front, to the observed driving behavior of the driver. The distance-speed characteristic curve assigns a target following distance to a number of speed values, which are referred to as support point speeds in the context of this description. The target following distance can, for example, be a time interval (hereinafter also referred to as a second interval), which indicates the time offset with which the vehicle should follow the vehicle in front. If the second interval is, for example, 1.5 seconds, the longitudinal guidance of the vehicle is controlled such that the vehicle always reaches a considered point (i.e.,a longitudinal position along the common lane of the vehicle and the vehicle in front). Alternatively, the target following distance can also be specified as a real distance, i.e., with a unit of length.

[0043] The characteristic curve can be adjusted in defined learning phases during a following journey influenced by the driver by overriding the distance control.

[0044] To define the learning phases, it is necessary to recognize as accurately as possible when the driver has set his desired distance. Road traffic is dynamic. New vehicles often merge, the vehicle accelerates, new speed limits appear, or sudden braking is necessary. In all of these situations, the driver deliberately does not drive at a constant distance behind a target object. Rather, the resulting distances can change dynamically and be temporarily larger or smaller than the actual desired distance. In one possible embodiment of the method proposed here or a corresponding learning algorithm, such situations should therefore preferably be ignored, i.e. the learning phases are deliberately scheduled outside of such driving situations.It is advantageous to limit the learning phases to situations of stable following with a comparatively constant following distance, in which it can be assumed that the driver has consciously adjusted his desired following distance.

[0045] With regard to the manner in which such meaningful following situations can be recognized, reference is made to the applicant's patent application "Method and System for Recognizing a Following Situation," filed with the German Patent and Trademark Office on January 15, 2024, the content of which is hereby incorporated into this application. The technical solutions described therein can be advantageously used within the framework of the method described here to define the learning phase(s).

[0046] In one step of the method, a previous (i.e., previously valid) target following distance is provided for the interpolation point speed of the distance-speed characteristic curve. This provision may, for example, involve reading the previous target following distance from a data memory in which various speed values with associated target following distances are stored.

[0047] In a further step, a recorded actual following distance is provided. The actual following distance can be recorded, for example, using a suitable sensor (e.g., radar) during the learning phase. As mentioned above, the learning phase can take place, for example, in a situation in which the driver consciously sets a desired following distance and follows a vehicle in front at a relatively constant desired following distance.

[0048] In a further step, a new (i.e., updated) target following distance for the interpolation point speed is determined based on the actual following distance. The new target following distance can also be determined based on the previous target following distance, for example.

[0049] Further possible influencing factors for determining the new target following distance are described in detail in the applicant's patent application filed with the German Patent and Trademark Office on January 15, 2024, entitled "Method for learning a following distance and system for automated longitudinal guidance of a vehicle." The content of that patent application is hereby incorporated into the present patent application.

[0050] According to one embodiment of the method proposed here, a method described in the above-mentioned patent application entitled “Method for learning a following distance and system for automated longitudinal guidance of a vehicle” is used to adapt the normal operating following distance as a function of the detected oversteer.

[0051] For example, it can be provided that the new target following distance is determined as a function of an override duration during which the automated longitudinal guidance is overridden. The override duration should not be understood as merely the duration of a specific accelerator pedal actuation by which the driver causes the automated longitudinal guidance to be overridden. Rather, the override duration refers to the entire duration during which, as a result of an override (which can also be caused, for example, by multiple accelerator pedal actuations spaced apart in time), a following distance is set that is different (in particular, reduced) than the following distance specified by the automated longitudinal guidance.

[0052] In particular, the new target following distance can be determined as a function of the actual following distance in such a way that the new target following distance is changed more compared to the previous target following distance, the longer the override duration is. For example, the new target following distance can be determined mathematically as a function of a factor that increases with the override duration during which the automated longitudinal guidance is overridden. In particular, it can be provided that the factor increases more than linearly with the override duration, for example with the square of the override duration. This is explained below using the example of a so-called motorway factor, which is used to calculate the new target following distance as a function of the actual following distance.

[0053] The motorway factor increases the longer a driver overrides the automated longitudinal guidance of the vehicle, especially on a motorway or motorway-like road. However, the restriction to overriding on a motorway or motorway-like road should be understood as merely optional in this context. Such a factor, which is included in the calculation of the new target following distance and increases with the duration of the override, can therefore, in principle, also be used during a learning process away from a motorway or motorway-like road.

[0054] Once the new target following distance has been determined, the previous target following distance in the distance-speed characteristic curve is replaced with the new target following distance in a next step. In this process, for example, a correspondingly updated distance-speed characteristic curve that assigns the new target following distance to the interpolation point speed can be saved.

[0055] The updated characteristic curve can, in particular, result in the adjusted normal operation following distance, which is to be adjusted in future normal operation of the automated longitudinal guidance at the corresponding speed.

[0056] A fourth aspect of the invention is a (data) processing device which has at least one processor and is configured to carry out the method according to the third aspect of the invention by means of the at least one processor. According to some embodiments, this may also be a spatially distributed processing device. For example, the processing device may comprise a plurality of spaced-apart sub-processing devices. The processing device used in the method according to the third aspect of the invention to set the normal operating following distance or the learning operating following distance to a preceding target object may be one of these multiple sub-processing devices, and the further processing device(s) used in the method may be one or more further sub-processing devices of the processing device according to the fourth aspect of the invention.

[0057] The processing device can, for example, comprise a control unit (or a part thereof), such as a control unit located on board the vehicle that controls the automated longitudinal guidance. However, embodiments are also possible in which the processing device is arranged entirely or partially outside the vehicle. It is conceivable, for example, that the processing device or parts of the processing device are part of a backend server to which the vehicle is connected via a wireless communication link.

[0058] According to an embodiment of the system according to the first aspect of the invention, the system comprises a processing device according to the fourth aspect of the invention or is data-technically connected to such a processing device - e.g. located on board the vehicle or in a backend.

[0059] A fifth aspect of the invention is a computer program comprising instructions which, when executed by a processing device (such as, for example, a processing device according to the fourth aspect of the invention), cause the processing device to execute a method according to the third aspect of the invention. The computer program can be divided into several separate subprograms, each of which can be executed by different, possibly spatially separated, processing devices (such as, for example, by several separate processors). A processing device according to the fourth aspect of the invention can therefore be configured, in particular programmed, to execute a computer program according to the fifth aspect of the invention.

[0060] A sixth aspect of the invention is a computer-readable storage medium comprising instructions that, when executed by a processing device, such as a processing device according to the fourth aspect of the invention, cause the processing device to execute a method according to the third aspect of the invention. In other words, a computer program according to the fifth aspect of the invention can be stored on the computer-readable storage medium.

[0061] The invention will now be explained in more detail using exemplary embodiments and with reference to the accompanying drawings. Fig. 1 illustrates, by way of example and schematically, a system for learning a following distance for automated longitudinal guidance of a vehicle. Fig. 2A illustrates exemplary and schematic steps of a method for learning a following distance for automated longitudinal guidance of a vehicle. Fig. 2B illustrates exemplary and schematically possible steps in adapting a normal operating following distance intended for a future following journey within the framework of the method from Fig. 2A. Fig. 3 illustrates an example of a distance-speed diagram with a distance-speed characteristic curve for a normal operation following distance and a distance-speed characteristic curve for a learning operation following distance. Fig. 4 illustrates an example of a normal operation following distance and a learning operation following distance in comparison with a normal distribution of a following distance desired by drivers.

[0062] The following describes the Fig. 1 schematically illustrated system 1 with reference to the Fig. 2A in the form of a block diagram, steps 40 to 44 of a method 4 for learning a following distance are explained.

[0063] The system 1 is configured to adapt a normal operating following distance N to a target object traveling ahead of the vehicle, which is provided for normal operation of the automated longitudinal guidance, depending on a detected override of the automated longitudinal guidance by a driver of the vehicle, so that a modified normal operating following distance is provided for a future following journey in normal operation of the automated longitudinal guidance.

[0064] The system 1 is further configured to set, depending on the situation, a learning operation following distance L to a target object in front that is increased compared to the normal operation following distance N, so that the driver has more opportunity to set a desired following distance that is reduced compared to the learning operation following distance L by overriding the automated longitudinal guidance.

[0065] For this purpose, the system 1 comprises a distance sensor system 12 for detecting the actual following distance of the vehicle from a vehicle in front (vehicle in front). For example, the distance sensor system 12 can have one or more radar sensors. Using a radar sensor, in addition to the following distance, a relative speed between the vehicle and the vehicle in front can also be determined, for example.

[0066] Furthermore, the system 1 comprises a speed sensor system 13 for detecting an actual speed of the vehicle. The speed sensor system 13 can comprise, for example, one or more wheel speed sensors, one or more acceleration sensors (from whose measured values an actual speed can be calculated), and / or a receiver for a global satellite navigation system (where an actual speed can be determined based on a change in the vehicle's position detected by the global satellite navigation system).

[0067] A (data) processing device 10 of the system 1 is signal-connected to the distance sensor system 12 and the speed sensor system 13 and is configured to receive a distance signal A comprising the actual following distance from the distance sensor system 12 and a speed signal G comprising the actual speed from the speed sensor system 13. The processing device 10 can also receive, for example, a relative speed signal V from the distance sensor system 12, from which, in combination with the speed signal G, a speed of the vehicle in front can be determined.

[0068] In the embodiment according to Fig. 1, the processing device 10 is further connected to an output device 16. The processing device 10 can generate control signals H for the output device 16 so that it outputs instructions or information to the driver. Fig. In Figure 1, the output device 16 is illustrated schematically and by way of example as a loudspeaker, since an indication can be output, for example, in acoustic form (e.g., as a voice output). Additionally or alternatively, the output device 16 can comprise, for example, a display for visual information display.

[0069] If a learning phase of a learning algorithm for learning a following distance is active, the driver can be informed of this fact via the output device 16. For example, the driver can be informed that their current override of the automated longitudinal guidance leads to a reduction of a previously stored normal operating following distance N.

[0070] In addition, the processing device 10 is signal-connected to a data memory 11 of the system 1 and is configured to receive characteristic curve information K from this, such as speed values and the desired following distance values assigned to them, in accordance with a distance-speed characteristic curve (hereinafter also referred to as characteristic curve) stored therein.

[0071] The processing device 10 is further configured to, in the manner described below with reference to Fig. 2B in the manner described in detail, depending on a recorded driving behavior (in particular depending on recorded actual distances and actual speeds), to determine updated characteristic curve information K' and to output this to the data memory 11 for storage, so that an updated characteristic curve is stored in the data memory 11 as a result.

[0072] The data memory 11 is signal-connected to a further processing device in the form of a control unit 14, which is designed to control the automated longitudinal guidance of the vehicle.

[0073] The control unit 14 is configured to set a normal operating following distance N to a target object traveling ahead during normal operation of the automated longitudinal guidance. The control unit 14 can, for example, read characteristic curve information K, K', which may in particular comprise a normal operating following distance for a current driving speed of the vehicle, from the data memory 11 and, depending on the characteristic curve information K, K', generate control signals S for controlling the automated longitudinal guidance, in particular to adjust the normal operating following distance N (as the target following distance) to a vehicle traveling ahead.

[0074] The control unit 14 can output the control signals S to a longitudinal guidance actuator 15, in particular to a braking system and / or to a drive module of the vehicle. The longitudinal guidance actuator 15 can then control the automated longitudinal guidance depending on the control signals S and, in particular, adjust a desired following distance to a vehicle in front. Fig. 1, the longitudinal guide actuator 15 is shown in dashed lines, since it can basically be a generic longitudinal guide actuator known per se in the prior art, which does not have to be understood as part of the system 1 according to the invention.

[0075] The system 1 is configured to use the control unit 14 and the processing device 10 to Fig. 2A in a schematic block diagram to execute steps 40-44 of the method 4 for learning a following distance.

[0076] In particular, the control unit 14 is configured to set a learning operation following distance L to a vehicle in front that is increased in comparison to the normal operation following distance N (step 41).

[0077] Furthermore, the processing device 10 is configured to detect an override of the automated longitudinal guidance by a driver of the vehicle, by which the following distance is shortened compared to the learning operation following distance L (step 42) and then to adapt the normal operation following distance N for a future following journey in normal operation of the automated longitudinal guidance depending on the detected override of the automated longitudinal guidance (step 43).

[0078] In the Fig. In the exemplary embodiment shown in Figure 1, the processing device 10, which carries out the aforementioned steps for detecting 42 an oversteering of the automated longitudinal guidance by a driver of the vehicle and for adjusting 43 the normal operation following distance depending on the detected oversteering, is a processing device separate from the control unit 14, which controls the automated longitudinal guidance of the vehicle and, in particular, also carries out the method step of setting 41 a learning operation following distance L. The control unit 14 is connected to the processing device 10 via signals and receives from it, for example, the distance and speed information A, G, V as the basis for the longitudinal control of the vehicle.

[0079] However, embodiments are also possible in which one and the same processing device carries out steps 41-43 of method 4. In other words, with reference to Fig. 1 the processing device 10 may be identical to the control device 14 or a part thereof (or vice versa).

[0080] The method 4 may further comprise, as an additional step 40 preceding steps 41-43, checking whether a switching condition is met. In Fig. In Figure 2A, this step is illustrated with a dashed box, as it is an optional additional step. In this case, it may be provided that the learning operation following distance L (instead of the normal operation following distance N) is only set if the check shows that the switching condition is met. For example, the switching condition may include the vehicle traveling on a specific road type, in particular on a highway or highway-like road.

[0081] In a further optional step 44 of method 4, the control unit 14 can, in a subsequent normal operation of the automated longitudinal guidance, generate control signals S for controlling the automated longitudinal guidance depending on the adjusted, updated normal operation following distance (namely, e.g., depending on correspondingly updated characteristic curve information K'), in particular to adjust the adjusted normal operation following distance or a target following distance to a preceding vehicle according to a correspondingly adjusted characteristic curve. The control signals S can be output to the longitudinal guidance actuator system 15.

[0082] Based on the adjusted normal operation following distance, the entire method 4 can be executed again in a later situation, with the modified normal operation following distance then taking on the role of the previous normal operation following distance. In other words, a learning operation following distance (possibly modified compared to the previous execution) that is greater than the previously modified normal operation following distance can be adjusted in order to give the driver an incentive to shorten the following distance by oversteering using the accelerator pedal. Depending on such oversteering, the previously modified normal operation following distance can then be further adjusted, so that the distance behavior desired by the driver can be gradually learned more and more accurately.

[0083] To adapt 43 the normal operation following distance, the processing device 10 of the system 1 can be set up to carry out the method steps 431 to 434 according to Fig. 2B to be executed.

[0084] Before the process steps 431 to 434 according to Fig. 2B are described in detail, should be considered with regard to the Fig. 3 briefly discusses an example of a distance-speed characteristic curve shown there.

[0085] In Fig. Figure 3 shows an example of a characteristic curve in the form of a diagram in which a second interval d is plotted against a speed v. An initially valid characteristic curve is shown in the diagram in the form of small circles connected by a solid line ("base" in the diagram legend). The data points represented as small circles each assign a second interval to a support point speed (hereinafter also referred to as "support point"). The characteristic curve between the support points results from a (here linear) interpolation.

[0086] In the example shown characteristic curve in Fig. 3 five sampling points are used. The larger the number of sampling points, the greater the personalizability of the characteristic curve. Therefore, it would generally be sensible to use significantly more than five sampling points, provided that the available memory allows this in a specific application. However, due to the interpolation that takes place anyway between the individual data points, increments of 1 km / h or even finer increments do not have to be selected between the sampling points in order to achieve a meaningful result. The most common speed limits such as 30, 50, 80, 100 and 120 km / h, as well as some higher speeds, e.g. 140, 160 or even 180 km / h, are sufficient to have a noticeable effect for the driver. If you add to this the speed limits of the system (i.e. in particular the automated driving function), you get, for example, 10 sampling points to be saved with 10 corresponding distance values to be saved.

[0087] The characteristic curve shown, for example, assigns a speed of approximately 56 m / s to a normal operating following interval N of just over 1.4 seconds, as shown in the Fig. 3 is illustrated by (finely) dashed auxiliary lines.

[0088] The illustrated (solid) characteristic curve, which in particular provides the normal operating following distance N for the speed of approximately 56 m / s, can be changed over time by a learning algorithm and thus adapted to the distance behavior desired by the driver.

[0089] According to an advantageous embodiment, the individual second intervals may not be taught completely freely, but certain restrictions can be provided within which they must move. In the diagram in Fig. 3, this is illustrated by a lower and an upper limit characteristic curve (see legend: "Limit"). The learned characteristic curve must be located between the upper and lower limit characteristic curves.

[0090] The upper limit can be determined, for example, by the performance of the sensor technology. A conventional radar sensor has a range of approximately 150 meters, with more advanced components even up to 300 meters. For objects further away, target object losses can occur, which can lead to longitudinal control behavior that the driver finds uncomfortable. If the distant control targets are repeatedly lost, this results in accelerations and braking that are incomprehensible to the driver. For example, in a following vehicle where the target object is traveling at 190 km / h, a following distance corresponding to the maximum radar visibility of 150 meters would correspond to a second interval of 2.84 s. To avoid incorrect braking or other undesirable effects, an additional buffer is preferably provided. Therefore, in the embodiment according to Fig. 3, a value of 2.5 seconds is assumed as the maximum interval per second, which represents the upper limit of the characteristic curve. Depending on the sensor performance, this can be further increased or decreased.

[0091] For the lower limit, legal requirements and standards can be considered as boundary conditions. For example, a distance resulting from the learned characteristic curve should not be less than the distance resulting from a characteristic curve that corresponds to the lowest distance level adjustable by the driver. For example, if an ACC function has three distance levels, the smallest distance level should form the lower limit. The lower limit characteristic curve in Fig. 3 corresponds to a limit thus defined.

[0092] In a system with three selectable distance levels, for example, a characteristic curve stored for the middle distance level can be used as a starting point for learning the characteristic curve. This characteristic curve can serve as a familiar reference point for the driver, who may have already driven previous models of the same vehicle brand. Learning in both positive and negative directions (with regard to the following distance) is quickly possible from this basis, so that a positive effect for the driver is noticeable immediately.

[0093] In the diagram in Fig. To illustrate the method 4 according to the invention, Figure 3 also shows a (roughly) dashed characteristic curve, which, for speeds higher than 10 m / s, provides a respective second interval as the learning operation follow-up interval, which is increased compared to the solid characteristic curve. For example, the (finely) dashed auxiliary lines illustrate that for a speed of approximately 56 m / s, a learning operation follow-up interval L is provided, which is approximately 1.6 seconds and is thus slightly less than 0.2 seconds longer than the normal operation follow-up interval N.

[0094] Method step 41, in which the learning operation following interval L is set, which is increased compared to the normal operation following interval N, can be carried out on the basis of such a modified characteristic curve. For example, if a switching condition of the type mentioned above is met, switching can be performed between the solid characteristic curve with the normal operation following interval N and the dashed characteristic curve with the learning operation following interval L, so that as a result, for example, at a speed of 56 m / s, the learning operation following interval L is set instead of the normal operation following interval N.

[0095] In the example shown, the dashed characteristic curve deviates from the solid characteristic curve only for comparatively high speeds of more than 10 m / s, because the inventive setting 41 of the increased learning mode following distance, as already mentioned, can be particularly relevant when driving on the motorway, where speeds of 10 m / s and less are normally not important.

[0096] Certain driving situations or time periods during a journey can be identified as suitable for learning the characteristic curve. This can particularly apply to a following journey that is relatively constant in terms of a following distance and is controlled (or at least influenced) by the driver. In such a situation, a learning phase can be started in which, at several points in time, e.g., in successive calculation cycles, the steps 431-434 described below are carried out in accordance with Fig. 2B can be executed.

[0097] In a step 431, a previous (ie previously valid) target following distance d n for an assigned support point velocity v n the previously applicable distance-speed characteristic curve. This provision can, for example, involve reading out the previous target following distance d n from the data memory 11, in which various speed values with associated target following distances are stored according to the characteristic curve.

[0098] In a further step 432, a detected actual following distance d ist The actual following distance d ist e.g. during a phase in which the driver overrides the automated longitudinal guidance, by means of the distance sensor system 12. The detected actual following distance d istreceived by the processing device 10 with the distance signal A from the distance sensor 12 and made available for further data processing.

[0099] In a further step 433, depending on the actual following distance d ist and the previous following distance d n a new (ie updated) target following distance d n , for the support point velocity v n certainly.

[0100] Subsequently, in a further step 434, the previous target following distance d n for the support point v n in the characteristic curve by the new target following distance d n , replaced. In this process, for example, an updated characteristic curve information K', according to which the support point v n the new target following distance d n , is assigned, are stored in the data memory 11, so that an updated characteristic curve is stored in the data memory 11 as a result.

[0101] When determining the new target following distance d n , in step 433, according to an embodiment described in detail below, at several times of the learning phase (e.g. in each calculation cycle) a respective new target following distance d n , based on the respective previously applicable (e.g. determined in a previous calculation cycle) target following distance d n and the current actual following distance d ist determined according to the following equation: dn'=dn+a∗fAB∗ft∗fδv∗fk∗(distAext−dn)

[0102] On the right side of the equation, there is a learning speed factor a, a highway factor f AB , a learning duration factor f t , a speed difference factor f δv , a plausibility factor f k and a factor which is determined by a deviation between the actual following distance d istand the previous target following distance d n depends on.

[0103] It should be noted that embodiments are also possible in which an expression for determining the new target following distance d n , - e.g. next to the previous target following distance d n - only one or some of the above-mentioned additional influencing factors are used.

[0104] In one possible embodiment, at least one of the above-mentioned influencing factors is taken into account, which is determined by a deviation between the actual following distance d ist and the previous target following distance d n depends, such as the last factor in the second term on the right-hand side of the above equation. Preferably, a speed difference factor f δv and / or a learning duration factor f t and / or a plausibility factor f k and / or a motorway factor f AB taken into account.

[0105] The above-mentioned possible influencing factors are described in detail in the applicant’s above-mentioned patent application entitled “Method for learning a following distance and system for automated longitudinal guidance of a vehicle”.

[0106] The motorway factor f AB explained in more detail as an example of a concrete implementation in which the determination 433 of the new target following distance is carried out as a function of an override duration during which the automated longitudinal guidance is overridden.

[0107] If a driver activates the adaptive cruise control system and exceeds the threshold several times without noticing a real change in the distance to the vehicle in front, the driver could prematurely deactivate the system or switch to a standard distance level. Therefore, it makes sense to provide a logic for this special case that enables significantly faster adaptation to the learning target.

[0108] This can be achieved by using the learning algorithm to determine the new target following distance d n , a motorway factor f AB which is greater the longer the automated longitudinal guidance is oversteered by a driver of the vehicle on a motorway or motorway-like road.

[0109] For a concrete design of the motorway factor f ABSeveral criteria need to be considered. On the one hand, this additional function should preferably only be active when the vehicle is on a highway or a highway-like road. In urban traffic, periods of oversteering may frequently occur, which do not necessarily count toward the distance target. Furthermore, the usage rate in urban traffic is not as high as on other road types.

[0110] In the present embodiment, the motorway factor f ABso called because it is only effective in a motorway environment. In general, however, such a factor does not necessarily have to be designed in such a way that its influence is limited only to situations in which the vehicle is traveling on a motorway or motorway-like road. Oversteering can also be taken into account on other road types in the manner described here when determining 433 a new target following distance, in particular in such a way that the new target following distance is changed more in comparison to the previous target following distance, the longer the detected oversteer duration is.

[0111] The influence of the motorway factor f AB should preferably be operated with an override duration t ueb , during which the automated longitudinal guidance is overridden. Since a large effect is necessary in long learning phases, a quadratic relationship is chosen. For example, the motorway factor fAB can be determined using the following equation: fAB=aAB⋅tueb2, tueb∈[0,tmax] where a AB a fixed value to normalize the time and t ueb is the duration of the violation, which for the purposes of algorithmic processing is included in the motorway factor f AB at a maximum value t max can be capped.

[0112] The duration of the violation t ueb can be used analogously to the formula for the learning duration t l which is described in the above-mentioned patent application entitled "Method for learning a following distance and system for automated longitudinal guidance of a vehicle". In other words, at the beginning of the learning phase, a dt0 and a v t0be set, but preferably with an additional time period at the beginning. This is helpful in actually setting the initial value to the correct one and avoiding any incorrect learning maneuvers. Otherwise, the additional factor may subsequently abort the adaptation acceleration despite only slight deviations. During this additional time period, the driver can adjust his or her desired distance by exceeding the limit. This takes a certain amount of time, while the vehicle must still accelerate accordingly in order to maintain the driver's constant desired distance. As soon as all conditions for learning are met, the driver actively overrides the accelerator pedal, and a determination of the road type actually results in a motorway or motorway-like road, a corresponding timer is started. Only when this timer expires will t ueb to 0, as well as d t0 and v t0\to the current values. The calculation of the motorway factor f then begins AB analogous to the formula for the learning duration t l .

[0113] The diagram in Fig. 4 illustrates an example of a normal operation following distance N and a learning operation following distance L in comparison with a statistical normal distribution of a following distance desired by drivers.

[0114] This illustration is intended to illustrate that the inventive increase in the following distance from a normal operating following distance to a learning operating following distance gives users of an automated longitudinal guidance function, in their statistical entirety, more opportunity to set a desired following distance by overriding the automated longitudinal guidance, whereby a desired distance behavior can be learned more effectively. Increasing the following distance to the learning operating following distance L is useful so that a correction, namely in particular a shortening of the following distance, can be made by the driver easily via the accelerator pedal. The initial increase in the following distance increases the probability of the driver requesting a reduction in the following distance.

[0115] If a Gaussian normal distribution is assumed for the desired following distance, shifting the distance characteristic (i.e., in particular, increasing the following distance to the learning mode following distance L) can have a large effect, as the diagram in Fig. 4 shows. Although more drivers will have to adjust their distance as a result, drivers who would otherwise only be able to adjust their desired distance by deactivating the system or the self-learning function are now also offered the option of using the self-learning system.

[0116] The vertical dashed lines mark the normal operation following distance N (left) and the learning operation following distance L (right). The curve itself represents the desired distances of the drivers. All users who are located to the left of the respective vertical dashed line in the normal distribution have the option of setting their desired distance by overstepping the accelerator pedal. The desired distance is less than that of the respective initial characteristic curve and can therefore be achieved via this route. All other users who are to the right of the respective vertical dashed line must deactivate the system or change the distance level manually. By shifting the distance characteristic curve, the number of users who can set their desired distance by overstepping the pedal is significantly increased.The learning algorithm is thus able to quickly adapt to individual preferences even for drivers who primarily use adaptive cruise control systems on highways. This option is also possible with active driver assistance, expanding the system's adaptability. QUOTES CONTAINED IN THE DESCRIPTION

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

[0000] DE 10 2015 016 993 A1

[0007] Cited non-patent literature

[0000] 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

[0006]

Claims

[1] System (1) for learning a following distance for automated longitudinal guidance of a vehicle, wherein the system (1) is configured to adapt a normal operation following distance (N) provided for normal operation of the automated longitudinal guidance to a target object traveling ahead of the vehicle as a function of a detected override of the automated longitudinal guidance by a driver of the vehicle, so that a modified normal operation following distance (N) is provided for a future following journey in normal operation of the automated longitudinal guidance, and wherein the system is further configured to set, depending on the situation, a learning operation following distance (L) to a target object traveling ahead that is increased compared to the normal operation following distance (N), so that the driver has more opportunity to set an actual following distance desired by him that is reduced compared to the learning operation following distance (L) by overriding the automated longitudinal guidance. [2] System (1) according to claim 1, wherein the system (1) is configured to temporarily set the learning operation following distance (L) and then gradually approximate the actual following distance to the target object driving ahead to the normal operation following distance (N) as long as the driver does not override the automated longitudinal guidance. [3] Vehicle with a system (1) according to one of the preceding claims. [4] Computer-implemented method (4) for learning a following distance for automated longitudinal guidance of a vehicle, wherein the vehicle comprises a processing device (14) which is configured to set a normal operating following distance (N) to a target object traveling ahead during normal operation of the automated longitudinal guidance, and wherein the method (4) comprises the following steps carried out by means of the processing device (14) and / or by means of one or more further processing devices (10): - setting (41) a learning operation following distance (L) to a target object traveling ahead that is increased compared to the normal operation following distance (N); - detecting (42) an oversteering of the automated longitudinal guidance by a driver of the vehicle, whereby an actual following distance is shortened compared to the learning operation following distance (L); and - adjusting (43) the normal operation following distance (N) for a future following journey in normal operation of the automated longitudinal guidance, wherein the adjustment of the normal operation following distance (N) takes place as a function of the detected override of the automated longitudinal guidance. [5] Method (4) according to claim 4, wherein the method (4) comprises, as a further step preceding said steps (41)-(43), checking (40) whether a switching condition is met, and wherein the learning operation following interval (L) is only set if the checking (40) shows that the switching condition is met. [6] Method (4) according to claim 5, wherein the switching condition comprises that the vehicle is driving on a certain type of road, in particular on a motorway or a motorway-like road. [7] Method (4) according to one of claims 4 to 6, wherein the adaptation (43) of the normal operation following distance comprises that for at least one support point speed of a distance-speed characteristic curve during a learning phase, the following steps are carried out: - providing (431) a previous target following distance for the support point speed; - providing (432) a detected actual following distance; - determining (433) a new target following distance for the support point speed as a function of the actual following distance; and - Replacing (434) the previous target following distance in the distance-speed characteristic curve with the new target following distance. [8] Method (4) according to claim 7, wherein the determination (433) of the new target following distance is carried out as a function of an override duration during which the automated longitudinal guidance is overridden. [9] Method (4) according to claim 8, wherein the determination (433) of the new target following distance is carried out as a function of the actual following distance in such a way that the new target following distance is changed more than the previous target following distance, the longer the override duration is. [10] Processing device (10, 14) having one or more processors, wherein the processing device is configured to carry out the method (4) according to one of claims 4 to 9 by means of the processor or by means of the plurality of processors. [11] Computer program comprising instructions which, when the computer program is executed by at least one processing device (10, 14), cause the latter to carry out a method (4) according to one of claims 4 to 9. [12] Computer-readable storage medium comprising instructions which, when executed by at least one processing device (10, 14), cause the latter to carry out a method (4) according to one of claims 4 to 9.

Images