Method and processing unit for controlling an automated driving function

Abstract

The invention relates to a method (2) for controlling an automated driving function of an ego vehicle (1), which driving function enables automated driving according to SAE level 3 or 4, the method comprising the following steps, which are carried out while the automated driving function is activated: detecting (21) sensor information relating to traffic behind the ego vehicle (1); determining (22) a distance (d) between the ego vehicle (1) and a vehicle (3) in the traffic behind; comparing (23) the determined distance with a predetermined safety distance (d0); if the determined distance (d) falls below the predetermined safety distance (d0) for a predetermined duration; and triggering (24) a request to the driver of the ego vehicle (1) to take over control of the vehicle.

Description

[0001] 24-2073 1

[0002] Method and processing device for controlling an automated driving function

[0003] The invention relates to a method for controlling an automated driving function of a vehicle, which enables automated driving according to SAE Level 3 or SAE Level 4. Furthermore, the invention relates to a processing device and a computer program for executing such a method, as well as a computer-readable storage medium on which such a computer program is stored.

[0004] Motor vehicles are often equipped with automated driving functions that enable automated longitudinal and / or lateral control. Some modern vehicles already have automated driving functions according to SAE Level 3 or even SAE Level 4.

[0005] SAE Level 3 refers to the SAE J3016 standard from SAE International as of April 30, 2021, and denotes a level of automation where the vehicle fully takes over the dynamic driving tasks, but the driver must still be constantly ready to take control.

[0006] SAE Level 4, according to the SAE J3016 standard, denotes an even higher level of automation, where, in certain defined environments or conditions, driving tasks are performed completely without human intervention. The driver can perform other activities or even sleep while driving, as the vehicle is able to navigate autonomously and react to unexpected events. However, the use of this technology is limited to specific areas or weather conditions where the system functions reliably.

[0007] One challenge with such automated driving functions is minimizing the risk of collisions with traffic behind. While modern vehicle sensors can detect objects in front of the vehicle very precisely, false detections of so-called "ghost objects" can still occur. In such cases, the vehicle system triggers sudden and strong braking maneuvers to avoid a supposedly imminent collision. The abrupt deceleration of the autonomous vehicle creates the risk that following traffic will not be able to brake in time, resulting in rear-end collisions. 24-2073 2

[0008] It is an object of the present invention to provide a method for controlling an automated driving function according to SAE level 3 or 4, which addresses the above-mentioned challenge regarding collision avoidance with rear traffic.

[0009] The problem is solved by the subject matter of the independent claims. Advantageous embodiments are specified in the dependent claims.

[0010] According to a first aspect of the present invention, a computer-implemented method for controlling an automated driving function according to SAE Level 3 or 4 is provided, wherein the method is executed while the automated driving function is activated.

[0011] A second aspect of the invention relates to a computer-implemented method for controlling an automated driving function according to SAE Level 3 or 4, which is executed while the automated driving function is not activated.

[0012] A vehicle (hereinafter also referred to as an ego-vehicle) capable of performing automated driving functions can, in particular, be a motor vehicle. The term "motor vehicle" here refers specifically to a land vehicle that is moved by mechanical power and is not bound to railway tracks. A motor vehicle in this sense can, for example, be a passenger car, motorcycle, or tractor.

[0013] Both methods include the steps of acquiring sensor information regarding rear traffic, determining a distance to a rear vehicle, and comparing this distance with a predetermined safety distance.

[0014] The predetermined safety distance can be chosen differently in the method according to the second aspect of the invention than in the method according to the first aspect of the invention. However, the same safety distance can also be used in both methods.

[0015] In the method according to the first aspect of the invention, if the safety distance is breached for a certain period of time, a request is triggered to the driver to take over control of the vehicle (so-called Take-over Request - TOR). 24-2073 3

[0016] In the method according to the second aspect of the invention, the activation of the automated driving function is suppressed if the determined distance falls below the predetermined safety distance. Thus, before the automated driving function is activated, it is checked whether the distance between a rear vehicle and the self-driving vehicle is sufficient, and if this is not the case, the system controlling the automated driving function is configured so that the function cannot be activated in the first place.

[0017] Control signals can be generated that configure the system so that the automated driving function cannot be activated – for example, by the driver. In particular, the activation of the automated driving function can be suppressed as long as the distance between a vehicle behind and the self-driving vehicle falls below the predetermined safety distance.

[0018] The acquisition of sensor information regarding rear traffic refers specifically to the acquisition of sensor information about an area behind the ego vehicle where rear traffic is generally expected. This acquisition can also occur even if no rear traffic is present. If rear traffic is present, for example, a single vehicle (the rear vehicle) can be detected, or multiple objects can be detected as rear traffic and provided, for example, in the form of an object list.

[0019] The acquisition of sensor information regarding rear traffic and the determination of the distance to the rear vehicle can be accomplished in a known manner using environmental sensors on the ego vehicle, such as a camera sensor, a radar sensor, a lidar sensor, and / or an ultrasonic sensor. In other words, the sensor information can include camera data, radar data, ultrasonic sensor data, and / or lidar data.

[0020] For example, a control unit that controls the automated driving function of the ego vehicle can receive information about the vehicle behind from the environmental sensors, such as an object list with distance information to one or more relevant objects behind the vehicle.

[0021] Regarding the predetermined safety distance as well as the measured (actual) distance between the rear vehicle and the first vehicle, it should be noted that these can be specified or determined directly as spatial distances, i.e., in the form of distances measured in meters or centimeters, and compared with each other. However, this need not be the case in all embodiments. Rather, the predetermined safety distance can also be defined as a temporal distance (so-called...), as is common practice in such contexts.

[0022] A distance of a few seconds can be specified, indicating the time a vehicle at that distance from the ego vehicle would need at its current speed to reach the current position of the ego vehicle in front. The actual distance determined could then also be expressed as a distance of a few seconds and compared to the safety distance.

[0023] To further enhance safety, the predetermined safety distance can, according to one embodiment, be dynamically adjusted to current driving conditions such as the speed of the ego vehicle and / or the vehicle behind. In particular, it can be provided that the safety distance is chosen to be larger the higher the speed of the ego vehicle and / or the higher the speed of the vehicle behind.

[0024] If, as described above, the predetermined safety distance is set as a second interval, such a dependency on the speed of the rear vehicle is automatically given, because with a fixed predetermined second interval, the corresponding "meter distance" is greater the greater the speed of the rear vehicle.

[0025] For example, the predetermined safety distance can be chosen as a interval of seconds, which is in the range of 1 to 2 seconds.

[0026] It is also possible to set the predetermined safety distance according to the well-known "half-speedometer" rule, i.e., the speed of the following vehicle in km / h is divided by two to obtain the safety distance in meters.

[0027] At higher speeds or in more critical situations, the safety distance can be increased accordingly to provide the driver with sufficient reaction time. Conversely, at lower speeds or in less critical situations, the distance can be reduced to increase driving comfort and efficiency. This allows the system to react flexibly to changing traffic conditions and always ensure an appropriate safety distance. 24-2073 5

[0028] The duration after which the prompt to the driver is triggered, according to the first aspect of the invention, can also be adapted to current driving conditions, such as the current speed of the first vehicle and / or the current speed of the second vehicle, according to one embodiment. In particular, the duration can be chosen to be shorter the higher the speed of the first vehicle and / or the higher the speed of the second vehicle.

[0029] It is also within the scope of the invention that the time until the gate is activated can be adjusted depending on how much the safety distance is breached. A longer time until the gate is activated can be permitted if the breach is only slight than if it is significant. In other words, the duration can be adjusted such that a shorter time is set for a comparatively significant breach of the predetermined safety distance than for a comparatively minor breach (and vice versa).

[0030] Thus, the duration can also depend on the measured distance itself, namely in particular in that the smaller the actual distance, the shorter the period for which a breach of the safety distance is tolerated (and vice versa).

[0031] For example, at higher speeds or in more critical situations (e.g., when the distance to the vehicle behind is very short), the system can warn the driver of the self-driving car earlier, allowing them to take control in time. In situations with lower risks, the warning can be issued later to avoid unnecessarily disturbing the driver.

[0032] Other possible influencing factors that can determine the choice of the predetermined duration (alone or in combination with several of the influencing factors) include:

[0033] - one or more weather conditions; prevailing visibility conditions;

[0034] - a time of day;

[0035] - a road condition;

[0036] - whether the first vehicle and / or the second vehicle are in a tunnel;

[0037] - a traffic situation, such as traffic density in the vicinity of the ego vehicle, and in particular whether there is a traffic jam;

[0038] - in which lane the ego vehicle is driving;

[0039] - a vehicle type and / or driving behavior of a vehicle driving in front of the ego vehicle. 24-2073 6

[0040] For example, it can be assumed that a camera used by the autonomous vehicle to detect objects it might have to brake for will have poor visibility at night or generally in the dark, in rain or fog, or when the sun is low, which can lead to an increased number of false positive object detections (ghost objects). In this case, it may be useful to trigger the handover request to the driver earlier, i.e., to specify a comparatively short predetermined time until the TOR (Traffic Gateway).

[0041] In tunnels, interference with the environmental sensors of the ego vehicle, especially radar sensors, is more likely than outside of tunnels, so false-positive object detections can occur more frequently here as well, and therefore a comparatively short predetermined duration until the TOR may be useful.

[0042] On a poor (e.g. uneven) road that has potholes or bumps, false positive object detections are more likely, so setting a comparatively short predetermined duration until the TOR may be advisable.

[0043] In heavy traffic, and especially in traffic jams, there is a greater likelihood of vehicles cutting in front of the ego vehicle compared to less dense traffic. Therefore, in such situations (i.e., high traffic density / congestion), it can be advantageous to allocate a comparatively short time to the TOR (Traffic Gateway).

[0044] If the ego vehicle is located in the outer left lane of a multi-lane roadway (or the outer right lane in the case of left-hand traffic), vegetation (e.g., bushes) next to the guardrail may be present, leading to an increased number of false-positive object detections. Therefore, a relatively short predetermined time until the TOR (Target Object Detection) can be advantageous in such cases.

[0045] Regarding the driving behavior of a vehicle in front of the ego vehicle, the time until TOR can be set longer if the vehicle in front drives relatively constantly than if the vehicle in front drives very dynamically.

[0046] With regard to the vehicle type of the vehicle in front, it may be possible, for example, to set a longer duration until the TOR (Traffic Activation Time) is reached if the vehicle in front is identified as a truck, compared to if it is identified as a car or motorcycle. Trucks generally brake less sharply and therefore usually drive more consistently (and possibly more anticipatorily) than the other vehicle types mentioned, so the risk of sudden, impulsive braking is generally lower when the vehicle in front is a truck.

[0047] According to one embodiment, the predetermined duration is chosen to be in the range of 3 to 60 seconds, in particular, for example, in the range of 20 to 30 seconds.

[0048] According to a further development of the first and / or second aspect of the invention, it is provided that a vehicle type of the rear vehicle is determined and that the predetermined safety distance and / or the predetermined duration are set depending on the determined vehicle type of the rear vehicle.

[0049] For example, the vehicle type can be determined, as is known per se, using a classification algorithm based on the sensor data acquired regarding the vehicle behind. Classification can be performed, for example, to determine whether the vehicle behind is a truck, a passenger car, or a motorcycle. This also applies to the embodiment mentioned above, in which the predetermined duration until the TOR (Traffic Out of Control) is set depending on the vehicle type of a vehicle driving in front of the ego vehicle.

[0050] The dependence of the time until the TOR on the vehicle type can, for example, include the fact that the time is chosen to be shorter for a motorcycle or a truck as the following vehicle than for a car, so that as a result, an insufficient following distance is tolerated for a shorter time for a motorcycle or truck than for a car.

[0051] Furthermore, when a car is behind, the safety distance can be chosen to be smaller than when a truck is behind, since a car exerts high deceleration on the road and braking distances of trucks are generally longer than those of cars.

[0052] The required safety distance for a motorcycle as a rear vehicle may also differ from that for a truck and / or car. For example, it can be taken into account whether the motorcycle is traveling centrally behind the vehicle being driven or whether it is traveling to the side and could therefore still swerve if necessary. In the latter case, the safety distance could accordingly be chosen to be smaller than if the motorcycle were positioned centrally behind the vehicle being driven. 24-2073 8

[0053] It is also conceivable that, particularly if the vehicle behind is identified as a truck, the predetermined safety distance could be chosen to be in the range of 30 to 40 meters. For example, a minimum safety distance of 30 to 40 meters could be specified specifically for trucks, even in embodiments where otherwise a time interval of one second (i.e., a speed-dependent safety distance) is used.

[0054] Further possible embodiments of the method according to the first aspect of the invention relate to the manner in which the driver is prompted. This can be done in a manner known per se, for example by means of visual, acoustic and / or haptic signals. It is also conceivable that the manner in which the driver is prompted is adapted to the respective situation, for example by means of more conspicuous visual, acoustic and / or haptic signals in the case of higher risks, such as a very short distance between the vehicle behind and the vehicle in front.

[0055] If the driver takes over driving the vehicle upon request or indicates their willingness to do so, for example by operating the steering wheel and / or pedals, the automated driving function can be deactivated and the journey can continue in a manual driving mode.

[0056] However, if the driver does not take over control of the vehicle even after repeated warnings, e.g., in the form of a warning cascade increasing in intensity, or if he is recognized as incapable of taking over control of the vehicle (e.g., as asleep), the self-driving vehicle can be automatically brought to a safe state using a so-called Minimal Risk Maneuver (MRM) by, for example, bringing it to a stop in its current lane (or, in right-hand traffic, after changing to the rightmost lane, or, in left-hand traffic, after changing to the leftmost lane).

[0057] Overall, the methods according to the invention enable an increase in the safety of automated driving functions according to SAE Level 3 or 4 by taking the collision risk with rear traffic into account and reacting accordingly. Through the intelligent consideration of rear traffic and the dynamic adaptation to the driving situation, the collision risk can be significantly reduced.

[0058] A third aspect of the invention is a (data) processing device configured to execute a method according to the first and / or second aspect of the invention. Accordingly, the preceding and following explanations of the methods according to the invention, as well as their possible embodiments, can be understood analogously for the processing device according to the invention, and vice versa.

[0059] The processing device can have at least one processor and be configured to carry out the method according to the first aspect of the invention by means of the at least one processor.

[0060] According to some embodiments, the processing unit can also be a spatially distributed processing unit (for example, across several processors or microcontrollers spaced apart from each other).

[0061] For example, the processing unit can be a control unit or part of a control unit of the vehicle. In particular, it can be a control unit for controlling the automated driving function.

[0062] A fourth aspect of the invention is a computer program comprising instructions that, when executed by a processing unit (such as a processing unit according to the third aspect of the invention), cause it to execute a method according to the first and / or second aspect of the invention. The computer program may be divided into several separate subprograms, each of which can be executed by different, possibly spatially separated (sub-)processing units (such as several separate processors).

[0063] A processing device according to the third aspect of the invention can be configured (i.e. programmed) to execute a computer program according to the fourth aspect of the invention.

[0064] A fifth aspect of the invention is a computer-readable storage medium containing instructions that, when executed by a processing unit (possibly distributed across multiple processors), cause it to execute a method according to the first and / or second aspect of the invention. In other words, a computer program according to the fourth aspect of the invention can be stored on the computer-readable storage medium.

[0065] The invention will now be explained in more detail with reference to exemplary embodiments and the accompanying drawings. 24-2073 10

[0066] Fig. 1A illustrates, by way of example and schematically, a scenario in which an ego vehicle equipped with an automated driving function according to SAE level 3 or 4 is driving on a road, with another vehicle (rear vehicle) driving behind the ego vehicle at a distance greater than a predetermined safety distance.

[0067] Fig. 1B illustrates, by way of example and schematically, a scenario in which the rear vehicle from Fig. 1A dips into the safety distance to the ego vehicle.

[0068] Figs. 2-3 illustrate block diagrams of a respective method for operating an automated driving function of a vehicle, which can be implemented in the example scenarios from Figs. 1A and 1B.

[0069] Figure 1A illustrates, schematically and by way of example, a scenario in which an ego-vehicle 1, equipped with an automated driving function according to SAE Level 3 or 4, is driving on a road. A rear vehicle 3 is driving behind the ego-vehicle 1 at a distance d that is greater than a predetermined safety distance dO. A safety zone behind the ego-vehicle 3, defined by the safety distance dO, is shown hatched in Figure 1A.

[0070] Figure 1B now shows the case in which the rear vehicle 3 from Figure 1A enters the safety distance dO to the ego vehicle 1 (and thus the hatched safety area shown in Figure 1B). This represents a critical situation, as the risk of collision with the traffic behind is now increased.

[0071] Figure 2 illustrates, by way of example and schematic, the steps of a first method 2 for operating an automated driving function, which can be applied in the scenarios shown in Figure 1A and 1B, assuming that the automated driving function is already active in these situations.

[0072] In contrast, Figure 3 shows, by way of example and schematically, the steps of a second method 4 for operating the automated driving function, which can be applied in the scenarios shown in Figures 1A and 1B, assuming that the automated driving function is not yet active in these situations. 24-2073 11

[0073] These methods 2, 4 can be executed by a data processing unit 10 in the form of a control unit of the ego vehicle 1, wherein the control unit 10 can be, for example, a control unit that also controls the automated driving function as a whole.

[0074] Both procedures 2, 4 begin with the acquisition 21, 41 of sensor information regarding the traffic behind the ego vehicle 1. Subsequently, the distance d between ego vehicle 1 and rear vehicle 3 is determined 22, 42 and compared with the predetermined safety distance dO 23, 43.

[0075] If, in the first procedure 2, the comparison 23 shows that the distance d falls below the safety distance dO for a predetermined duration, as shown in Figure 1 B, in a further step 24 a TOR is triggered to the driver of the ego vehicle 1, i.e. a request to take over driving.

[0076] As an additional measure to reduce the risk of a rear-end collision, it may be provided that the driver of the rear vehicle 3 is alerted to the short distance and the resulting danger by activating light signals (e.g. hazard warning lights and / or brake lights) of the ego vehicle 1.

[0077] In the second method 4, it is provided that if the comparison 43 shows that the distance d falls below the predetermined safety distance dO (see Fig. 1 B), the activation of the automated driving function is suppressed. For example, it may be provided that the automated driving function cannot be activated until it is determined that no vehicle is (or is no longer) located within the safety zone behind the ego-vehicle 1 defined by the safety distance dO.

[0078] For both methods 2, 4, if the comparison 23, 43 shows that the distance d does not fall below the safety distance dO (in the case of the first method 2 for the predetermined duration), as illustrated in Figure 1A, the respective method 2, 4 starts again in step 25 or 45 with the renewed acquisition 21, 41 of the sensor information regarding the rear traffic, etc. This continuous monitoring of the distance to the rear vehicle 3 ensures that the system can react to critical situations at any time.

[0079] The predetermined safety distance dO can be chosen in both methods 2, 4 depending on various factors, such as the speed of the ego vehicle 1 24-2073 12 and / or the rear vehicle 3, a detected vehicle type of the rear vehicle 3, the weather conditions, the road surface or a reaction time of the driver.

[0080] The higher the speed or the worse the road conditions, the greater the safety distance dO should be chosen in order to be able to react in time in the event of a dangerous situation.

[0081] Similarly, the predetermined duration for which the distance d must fall below the safety distance dO in order to trigger a takeover request to the driver in the second procedure 2 can be determined depending on one or more of these or further factors and / or depending on the actual measured distance d. For example, a longer duration until the takeover request can be allowed if the safety distance dO is only slightly undershot than if it is significantly undershot, i.e., if the distance d is considerably smaller than the distance dO.

Claims

-2073 13 Patent claims 1. Method (2) for controlling an automated driving function of an ego vehicle (1) which enables automated driving in accordance with SAE Level 3 or SAE Level 4, wherein the method (2) comprises the following steps which are performed while the automated driving function is activated: Acquisition (21) of sensor information regarding rear traffic behind the ego vehicle (1); Determine (22) a distance (d) between the ego vehicle (1) and a rear vehicle (3) in reverse traffic; - Comparing (23) the determined distance (d) with a predetermined safety distance (dO); - if the specified distance (d) falls below the predetermined safety distance (dO) for a predetermined duration, triggering (24) a request to the driver of the ego vehicle (1) to take over vehicle control.

2. Method (2) according to claim 1, wherein the predetermined duration is dynamically adapted to current driving conditions, in particular to a current speed of the ego vehicle (1) and / or to a current speed of the rear vehicle (3) and / or to the determined distance (d).

3. Method (2) according to claim 2, wherein the duration is adjusted in such a way that the duration is chosen to be shorter the greater the speed of the ego vehicle (1) and / or the greater the speed of the rear vehicle (3).

4. Method according to claim 2 or 3, wherein the duration is adjusted in such a way that, in the case of a greater shortfall in the predetermined safety distance (dO), the duration is set shorter than in the case of a lesser shortfall in the predetermined safety distance (dO).

5. Method (2) according to any of the preceding claims, wherein the predetermined duration is selected depending on one or more of the following circumstances: - one or more weather conditions; - Visibility conditions; - a time of day; -2073 14 - a road condition; - whether the ego vehicle (1) and / or the rear vehicle (3) are in a tunnel; - a traffic situation, in particular whether there is a traffic jam; - in which lane the ego vehicle (1) is driving; - a vehicle type and / or driving behavior of a front vehicle driving in front of the ego vehicle (1).

6. Method (4) for controlling an automated driving function of an ego vehicle (1) which enables automated driving in accordance with SAE Level 3 or SAE Level 4, wherein the method (4) comprises the following steps which are performed while the automated driving function is not activated: Acquisition (41) of sensor information regarding rear traffic behind the ego vehicle (1); Determine (42) a distance (d) between the ego vehicle (1) and a rear vehicle (3) in reverse traffic; - Comparing (43) the determined distance (d) with a predetermined safety distance (dO); - if the specified distance (d) falls below the predetermined safety distance (dO), suppression (44) of the ability to activate the automated driving function.

7. Method (2, 4) according to one of the preceding claims, wherein the predetermined safety distance (dO) is dynamically adapted to current driving conditions, in particular to a current speed of the ego vehicle (1) and / or to a current speed of the rear vehicle (3).

8. Method (2, 4) according to claim 7, wherein the adjustment of the safety distance (dO) is carried out in such a way that the safety distance (dO) is chosen to be larger the greater the speed of the ego vehicle (1) and / or the greater the speed of the rear vehicle (3).

9. Method (2, 4) according to one of the preceding claims, wherein a vehicle type of the rear vehicle (3) is determined and wherein the predetermined safety distance and / or the predetermined duration is set depending on the vehicle type of the rear vehicle (3). -2073 15 10. Processing device (10) configured to carry out a method (2, 4) according to any one of the preceding claims.

11. Computer program comprising instructions which, when the computer program is executed by a processing device (10), cause it to execute a method (2, 4) according to any one of claims 1 to 9.

12. Computer-readable storage medium comprising instructions which, when executed by a processing unit (10), cause it to execute a method (2, 4) according to any one of claims 1 to 9.

Images