Method and system for predicting a functional quality of a driving assistance function

Abstract

The invention relates to a method for predicting the quality of a function of a driving assistance function, comprising the following steps: detecting (31), by a first vehicle (1), information characterizing the quality of a driver assistance function or related to the quality of the driver assistance function; determining (32), from the detected information and by means of a calculation device (61), a predicted quality of a function of a driver assistance function for a section of road of a second vehicle (2), which is identical to or different from the first vehicle (1); outputting (33), by means of an output device (21), information about the predicted quality of the function in a manner perceptible to a vehicle occupant of the second vehicle (2). Here, the detection (31) of the information takes place by accessing a control system (10) of the first vehicle (1); and / or the predicted quality of the function is determined (32) by accessing data provided by a control system (20) of the second vehicle (2).

Description

Technical Field

[0001] This invention relates to a method and system for predicting the functional quality of driving assistance functions. Background Technology

[0002] Various driver assistance systems (FAS) are known in the prior art, which can improve the safety and / or comfort of the driver or vehicle occupants. For example, driver assistance systems can take over certain driving tasks or even autonomously control the vehicle, within the scope of partially automated driving (TAF), highly automated driving (HAF), or fully automated driving (VAF).

[0003] For example, lane assist systems can be set up to keep the vehicle between lane markings. These markings can be scanned and automatically recognized, for example, using cameras. Other driver assistance systems do not directly intervene in guiding the vehicle but only provide certain signals, such as warning signals, thereby facilitating safe driving.

[0004] Examples of driver assistance systems that automatically intervene in vehicle guidance are longitudinal guidance assist systems, lateral guidance assist systems, and assist systems that couple longitudinal and lateral guidance. Known longitudinal guidance assist systems include, for example, automatic speed limiting, automatic speed control (such as the so-called Dynamic Cruise Control - DCC in BMW Group vehicles), or adaptive speed control (Adaptive Cruise Control or Active Cruise Control - ACC), etc. Lateral guidance assist systems, longitudinal guidance assist systems, or coupled steering and lateral guidance assist systems can, for example, implement so-called hands-free or feet-free functions, which alleviate some of the driver's steering or pedal operation tasks.

[0005] This type of driver assistance system is known to have multiple operating modes, which sometimes allow the driver to "take over." For example, it can be configured that the driver can take over the ACC function via the accelerator pedal and thus temporarily take over longitudinal guidance. If the driver subsequently removes their foot from the accelerator pedal again, the ACC function will continue to adjust to the set speed or the vehicle in front.

[0006] In addition, parking control assistants and automatic reversing assistants are known, the latter of which allows the storage (of a specific segment of road that has been passed at low speed) and automatic re-driving through that segment when needed.

[0007] Examples of driver assistance systems that automatically provide warning functions in certain situations (and may also include steering intervention and / or longitudinal guidance intervention in some forms) are lane change assist systems and assist systems for forward collision warning, pedestrian warning, side collision warning, intersection warning or lane departure warning.

[0008] In this document, a driver assistance system is generally understood to be a system designed, by means of software and / or hardware implementation, to output signals, such as warning signals, that are relevant to the driver or vehicle occupants in terms of safety or comfort, and / or to intervene in the control of the vehicle, or—permanently or at least temporarily—to take complete control of the vehicle, based on data acquired by one or more sensors.

[0009] Driver assistance systems (FAS) are typically deactivated if certain conditions for their operation are no longer met. These operating conditions, which must be met for safe and reliable FAS use, may be related to the current vehicle environment, particularly the road conditions. Therefore, once excessive steering torque is required to maintain the trajectory during sharp turns, such as when steering and lane guidance assist are discontinued. This situation occurs frequently and repeatedly, for example, on highway ramps. Other situations where FAS can typically be deactivated include crossing intersections and driving through roundabouts.

[0010] Typically, a take-over request (TOR) is issued to the driver before the FAS is deactivated (sometimes referred to as "offline"). This means that the driver is prompted in a way that is perceptible to the driver (i.e., audibly and / or visually) that he will soon have to manually take over control of the vehicle at least partially in order to prepare for the safe deactivation of the driver assistance system or specific driver assistance function of the FAS.

[0011] Driver satisfaction typically depends on the availability and performance of their vehicle's FAS (Fueling Autopilot). Drivers often feel uncomfortable when the FAS is frequently turned off, especially when they are surprised that the FAS is disabled, as they may have to manually intervene in the vehicle's steering and / or longitudinal guidance for short periods and with considerable effort. Summary of the Invention

[0012] The object of this invention is to provide a method and system that at least partially overcomes the shortcomings of the prior art.

[0013] The first aspect of the present invention relates to a method for predicting the functional quality of a driving assistance function.

[0014] The first step of the method according to the invention is to detect, via a first vehicle, information characterizing the functional quality of the driving assistance function or related to the functional quality of the driving assistance function.

[0015] In the context of this document, a vehicle is essentially any type of means of transport that can be used to move people and / or goods. Examples include: motor vehicles, trucks, land vehicles, buses, taxis, cable car cabins, elevator cabins, rail vehicles, vessels (e.g., ships, small boats, submarines, diving bells, hovercraft, hydrofoils), and aircraft (airplanes, helicopters, ground effect vehicles, airships, balloons).

[0016] Vehicles can specifically be motor vehicles. In this sense, a motor vehicle is a land vehicle that is powered by an engine and is not bound by tracks. In this sense, a motor vehicle can be designed as, for example, a car, a motorcycle, or a tractor.

[0017] The first vehicle is preferably equipped with driver assistance functions.

[0018] The method described herein can also provide multiple such first vehicles, preferably the entire convoy.

[0019] The detected information may include, for example, empirical data regarding the quality of the driving assistance function experienced by the first vehicle. This could be due to factors such as weather, ambient light, potential sensor glare, road curvature, or vehicle speed, which can affect the performance or availability of the driving assistance function. The detected information may also include, in particular, data from the first vehicle's environmental sensors (e.g., weather or speed data) and / or data from an environmental model used by the first vehicle. In other words, the first information can directly or indirectly describe the situation-dependent quality of the driving assistance function, especially its availability. Furthermore, the first information may include vehicle data of the first vehicle (e.g., data related to the hardware and / or software status).

[0020] A further step is to determine the predicted functional quality of the second vehicle's driver assistance functions for a specific road segment based on the detected information and with the aid of a computing device. The second vehicle may be the same as (i.e., it can be the first vehicle) or different; that is, it may be a different vehicle from the first vehicle.

[0021] A further step is to output information about the predicted functional quality in a manner perceptible to the occupants of the second vehicle via an output device. For example, this information may be output via a visual display and / or an audio message.

[0022] Information regarding the predicted functional quality is preferably output using a map that includes a graphical display of road segments.

[0023] In one possible implementation, the display device is located in or on the second vehicle. The display device may be, for example, a screen or projection device located in the cockpit, such as in the form of a touchscreen or head-up display. It may also be, for example, a display device for the onboard computer of the second vehicle, which, in addition to displaying data, can also display other information as needed, such as current speed, current RPM, or navigation instructions. For example, it may be specified that the driver can access the visual information through corresponding menus, within a software-implemented portal, and / or in applications and / or widgets and / or applets.

[0024] However, alternatively or additionally, a display device removable from or capable of being removed from the second vehicle may also be provided, for example, in the form of a mobile device. In other words, information about the predicted functional quality may be displayed, for example, on a mobile device such as a smartphone or laptop, or by a desktop computer located outside the second vehicle. For example, the driver may access this information via a special vehicle application on their mobile device or through a portal using a customer account. Thus, a mobile device or desktop computer can be used as a (potentially additional) display device for the usage data.

[0025] Information regarding the predicted quality of the function can, in particular, include predictions about whether the driver assistance function will (likely) be available on the relevant road sections. For the driver of the second vehicle, the advantage is that he can gain a consistent impression of the function's availability and can easily adapt to it in advance. Furthermore, the driver can gradually expand his consistent impression of the function's availability using the method according to the invention, and learn it better over time. In this way, unpleasant surprises, such as short-term deactivation of the driver assistance function, can be avoided. This information can also include, for example, positive recommendations indicating that the driver assistance function (likely) works particularly well on the relevant road sections.

[0026] According to this method, information is detected by accessing the control system of a first vehicle, wherein, according to some embodiments, the driver may optionally make manual input. The control system may, in particular, be a control system that controls the driving assistance functions of the first vehicle. The term "control system" should be understood herein to include, for example, corresponding logical computing mechanisms, such as one or more microcontrollers or processors, but may also include environmental sensors where necessary.

[0027] Alternatively or additionally, the method specifies that the predicted functional quality is determined by accessing data provided by the control system of the second vehicle, based on the characteristics upon which the information is based, through access to the control system of the first vehicle.

[0028] Preferably, the predicted functional quality is determined by considering information related to at least one of the following: the road route in the road segment, such as road curvature, represented, for example, by digital map information and / or by data detected using environmental sensors of the second vehicle; the software state of the second vehicle (the control system of the second vehicle, particularly related to driver assistance functions); the hardware of the second vehicle; the environmental model provided by the control system of the second vehicle; and information detected by the environmental sensors of the second vehicle, such as weather conditions, brightness, or glare. Furthermore, when determining the predicted functional quality, planned speeds and / or speed limits, for example, based on digital map information or traffic sign recognition in the road segment, may also be considered.

[0029] In an advantageous implementation, information regarding the predicted functional quality is provided to the control system of the second vehicle. Specifically, the method may include (automatically) controlling the driving assistance functions of the second vehicle based on the predicted functional quality. Controlling the driving assistance functions means selectively influencing them, such as (possibly partially) enabling, (possibly partially) disabling, limiting, or parameterizing them, where this can be specifically targeted with reference to relevant road segments.

[0030] As described above, and also within the scope of this invention, if it is determined that the quality of the function (e.g., one or more predetermined criteria, such as those affecting the safety or reliability of the driving assistance function) is insufficient, the method may further include automatically disabling the driving assistance function of the second vehicle before driving on the road segment.

[0031] According to one implementation, the detected information is stored and processed in a backend located at a distance from the first and second vehicles. The backend may, for example, include one or more computing units and one or more storage units. Thus, it may be stipulated, for example, that the detected information—preferably in an anonymous form—is transmitted to a backend server located outside the first and second vehicles. The backend may, for example, be operated by the vehicle manufacturer. In the backend, for example, over time, increasingly more information related to the prediction of the functional quality of the driver assistance functions (e.g., based on detected information from the entire fleet) may be aggregated. For example, a self-learning algorithm based on this can make increasingly reliable predictions of functional quality. Such a self-learning algorithm may, for example, be classical deterministic and / or based on one or more neural networks.

[0032] According to a second aspect of the invention, a system for predicting the functional quality of a driving assistance function is provided. The system includes a computing unit designed to: receive information characterizing the functional quality of the driving assistance function or information related to the functional quality of the driving assistance function, wherein the information has been detected using a first vehicle; and, based on the detected information, determine the predicted functional quality of the driving assistance function of a second vehicle (identical to or different from the first vehicle) for a given road segment. Furthermore, the system includes an output device for outputting information regarding the predicted functional quality in a manner perceptible to the occupants of the second vehicle.

[0033] The method according to the first aspect of the invention can be performed, for example, by a system according to the second aspect of the invention. Therefore, the description of the inventive method according to the first aspect of the invention is also applicable in a corresponding manner to the system according to the second aspect of the invention, and vice versa. Similarly, advantageous embodiments of the method according to the invention correspond to advantageous embodiments of the system according to the invention described herein, and vice versa.

[0034] According to one embodiment, the system according to the invention includes a rear end disposed outside a first vehicle and a second vehicle. The rear end may include the system's computing mechanism. Furthermore, the rear end may include a storage mechanism for storing detected information and / or information derived therefrom, such as information regarding the predicted functional quality of driving assistance functions.

[0035] The system may also include a transmission device for transmitting detected information from the first vehicle to the back end and / or a transmission device for transmitting information about predicted functional quality from the back end to the second vehicle. Data transmission is preferably performed wirelessly (e.g., via mobile radio or WiFi). Thus, to transmit relevant information from the first vehicle to the back end or from the back end to the second vehicle, a mobile radio interface according to 3G, 4G, or 5G mobile radio standards may be used, for example.

[0036] As described above, some embodiments of the method or system according to the invention achieve a self-learning predictive function for the functional quality of driver assistance functions. For example, data regarding function deactivation, including the current position, direction, speed, and acceleration of each vehicle, as well as other environmental parameters (e.g., weather, time, date, etc.) if necessary, are collected in the backend from the fleet via wireless communication. Based on this information, for example, a map is continuously and in real-time created, containing spatial, temporal, and other contextual parameters that the driver assistance function cannot control. This is done centrally in the backend, using input from the entire fleet of vehicles with the appropriate equipment. The functional availability model learned in this way can be wirelessly provided to each individual vehicle in the fleet, where the driver assistance function can be deactivated early, conveniently and transparently, if necessary. For example, if some vehicles in the fleet have previously given up at the apex of a highway ramp, automatic deactivation can occur well before a highway ramp. In this way, the driver can obtain a consistent impression of functional availability and easily adapt to it.

[0037] Another advantage is that the vehicle's behavior can dynamically adapt to changes (such as road routes or the state of its own hardware or software).

[0038] The method according to the invention can be applied to a wide range of driving assistance functions, such as steering and lane guidance assist, ACC, and parking assist. Attached Figure Description

[0039] The invention will now be explained in more detail with reference to embodiments and the accompanying drawings. Without departing from the scope of the invention, features and combinations thereof mentioned in the specification and / or shown separately in the drawings may be used not only in their respective specified combinations, but also in other combinations or individually.

[0040] Figure 1 A system for predicting the functional quality of a driving assistance function is illustrated schematically and exemplary.

[0041] Figure 2 A schematic flowchart of a method for predicting the functional quality of a driving assistance function according to an exemplary embodiment is shown.

[0042] Figure 3 It shows Figure 2 A schematic flowchart illustrating the improvement of the method. Detailed Implementation

[0043] Figure 1The example scenario involves two vehicles 1 and 2, wherein method 3 according to the invention is performed using system 100 for predicting the functional quality of driving assistance functions. The invention is explained below by way of this example scenario, wherein reference is also made to the method according to... Figure 2 and 3 The accompanying drawings, which illustrate exemplary flowcharts of exemplary embodiments of method 3, show steps 31-34.

[0044] Figure 1 The first vehicle 1 is shown, which is equipped with driving assistance functions such as steering and lane guidance assistants.

[0045] During driving, the control system 10 of the driving assistance function of the first vehicle 1 is accessed to detect information characterizing the functional quality of the driving assistance function and / or information related to the functional quality of the driving assistance function. Figure 2 and Figure 3 (Step 31 in the process). The information detected includes the quality of the driving assistance function experienced by the first vehicle 1 and / or empirical data on conditions related to the quality of the function (e.g., weather-related conditions).

[0046] For example, it can be detected whether a driver assistance function is available at a specific time during a particular curve in which the first vehicle 1 is traveling, or whether it is completely or partially deactivated (i.e. abandoned, for example, because the curve is too sharp at the driving speed used for automatic lateral guidance). Furthermore, the detected information may include other parameters such as weather conditions, driving speed, or the software status of vehicle 1.

[0047] The detected information is transmitted from the first vehicle 1 to the backend 6 via a mobile radio connection. The backend 6 includes a computing unit 61 and a storage unit 62, which are capable of processing or storing the detected information. The computing unit 61 receives the information detected by the first vehicle 1. The storage unit 62 may, for example, serve as a memory or temporary storage for the detected information or information thus derived therefrom.

[0048] In addition to the information detected by the first vehicle 1, many other vehicles (not shown) can also provide corresponding information to the backend 6. In other words, the information detected in the backend 6 can be aggregated from the entire fleet.

[0049] Based on the detected information, the computing unit 61 in the backend 6 determines the predicted functional quality of the driving assistance function of the second vehicle 2, which differs from that of the first vehicle 1, for a specific road segment A. Figure 2 and Figure 3 (Step 32 in the text). Road segment A could be, for example, the aforementioned curve that the first vehicle 1 has passed through. In particular, road segment A could be the road segment that the second vehicle 2 is about to travel according to the current route plan.

[0050] Information regarding the quality of predicted driver assistance functions, as determined in this way, may in particular include predictions about whether the driver assistance function is available on road segment A. For example, computing unit 61 may use a self-learning algorithm to provide increasingly reliable predictions about the availability of driver assistance functions over time, based on information detected by the fleet.

[0051] Information regarding the predicted functional quality of the driver assistance function is transmitted from the backend 6 to the second vehicle 2 via a cellular connection, and output to the occupants of the second vehicle 2 via an output device 21. Figure 2 and Figure 3 Step 33 in the middle.

[0052] In this embodiment, the output device 21 includes a screen that graphically displays road segment A and marks it with shaded areas. This marking might, for example, indicate that, under given conditions (e.g., current weather, driving speed, and the software state of the driver assistance functions of the second vehicle 2), the driver assistance functions are predicted to be unavailable on road segment A. As a result, the driver of the second vehicle 2 is aware in advance that a takeover request will be issued before road segment A.

[0053] Information regarding the predicted functional quality can also be directly provided to the control system 20 of the driving assistance functions of the second vehicle 2. In this case, according to Figure 3 A variant of the method schematically illustrated could include an additional step 34 in which the driving assistance functions of the second vehicle 2 are controlled based on the predicted quality of functionality. This means, for example, that the driving assistance functions can be automatically enabled, disabled, limited, or parameterized based on the predicted quality of functionality, using the control system 20. Specifically, if determination 32 indicates that the predicted quality of functionality is insufficient, the driving assistance functions of the second vehicle 2 can be automatically disabled before traveling on road segment A. For example, if the predicted quality of functionality already presupposes abandoning the driving assistance functions in the area of ​​road segment A, automatic disabling can be implemented earlier if necessary. This avoids the driver being unpleasantly surprised by the short-term disabling of the driving assistance functions.

[0054] The determination of the predicted functional quality 32 can be performed by accessing (and considering) data provided by the control system 20 of the second vehicle 2. This data may, for example, relate to information about the software and / or hardware status of the second vehicle 2. Furthermore, this data may relate to information about the environmental model provided by the control system 20 of the second vehicle 2. Additionally, this data may include information detected by environmental sensors of the second vehicle 2 (e.g., information about weather, brightness, road direction, traffic signs, etc.).

[0055] In the embodiments described herein, the second vehicle 2 is different from the first vehicle 1. However, in other embodiments, the second vehicle 2 may also be the same as the first vehicle 1. In this case, the detected information may be stored and processed locally in the vehicle, for example, without involving the backend. While the vehicle cannot benefit from knowledge obtained from other vehicles, it can collect increasingly reliable information about the predicted functional quality of the driver assistance functions over time due to the data detected by the vehicle itself, which is beneficial to the driving experience of the vehicle occupants. This design variant is no different from the design variant with two different vehicles 1 and 2 in terms of the display and further use of information about the predicted functional quality, and therefore reference can be made to the above description.

Claims

1. A method (3) for predicting the functional quality of a driving assistance function, wherein the driving assistance function needs to be deactivated and a takeover request is issued to the driver of the vehicle if the functional quality is insufficient, the method comprising the following steps: - Information characterizing the functional quality of the driving assistance function or related to the functional quality of the driving assistance function is detected (31) by the first vehicle (1), wherein the detection (31) of the information is performed by accessing the control system (10) of the first vehicle (1); - Based on the detected information and with the aid of a computing unit (61), determine (32) the predicted functional quality of the driving assistance function of the second vehicle (2) for road segment (A), the second vehicle being the same as or different from the first vehicle (1), wherein the predicted functional quality is determined (32) by accessing data provided by the control system (20) of the second vehicle (2); - Information about the predicted functional quality is output (33) via the output device (21) in a manner perceptible to the occupants of the second vehicle (2); - If the determination (32) indicates that the predicted quality of the function is insufficient, the driving assistance function of the second vehicle (2) is automatically deactivated before driving on the road segment (A) to avoid issuing a takeover request to the driver of the second vehicle (2) while driving on the road segment (A).

2. The method (3) according to claim 1, wherein, The output (33) of information regarding the predicted functional quality is made using a graphically displayed map, which includes the road segment (A).

3. The method (3) according to claim 1 or 2, wherein, Information regarding the predicted quality of the function includes predictions about whether the driving assistance function will be available on the road segment (A).

4. The method (3) according to claim 1 or 2, wherein, The detected information is stored and processed in the rear end (6) at a certain distance from the first vehicle (1) and the second vehicle (2).

5. The method (3) according to claim 1 or 2, wherein, The information detected includes empirical data on the quality of the driving assistance function experienced by the first vehicle (1).

6. The method (3) according to claim 1 or 2, wherein, The determination (32) of the predicted functional quality is made taking into account information relating to at least one of the following: - The road route in the aforementioned road segment (A); - The software status of the second vehicle (2); - Hardware of the second vehicle (2); - An environmental model provided by the control system (20) of the second vehicle (2); - Information detected by the environmental sensors of the second vehicle (2).

7. The method (3) according to claim 1 or 2, wherein, Information about the predicted functional quality is provided to the control system (20) of the second vehicle (2).

8. The method (3) according to claim 1 or 2, further comprising: - Control (34) the driving assistance functions of the second vehicle (2) based on the predicted quality of the functions.

9. A system (100) for predicting the functional quality of a driving assistance function, wherein the driving assistance function needs to be deactivated and a takeover request issued to the driver of the vehicle if the functional quality is insufficient, the system comprising: - The computing mechanism (61) is designed for: Receive information characterizing the functional quality of the driving assistance function or related to the functional quality of the driving assistance function, wherein the information has been detected by the first vehicle (1); Based on the detected information, determine (32) the predicted functional quality of the driving assistance function of the second vehicle (2) for road segment (A), wherein the second vehicle is the same as or different from the first vehicle (1); If the determination (32) indicates that the predicted quality of the function is insufficient, the driving assistance function of the second vehicle (2) is automatically deactivated before driving on the road segment (A) to avoid issuing a takeover request to the driver of the second vehicle (2) while driving on the road segment (A).

10. The system (100) according to claim 9, further comprising: - Output device (21) for outputting (33) information about the predicted functional quality in a manner perceptible to the occupants of the second vehicle (2).

Images