Method and processing device for controlling an automated driving function

Abstract

A computer-implemented method (2) for controlling the automated driving function of a vehicle comprises the steps: receiving (21) environmental sensor data from several environmental sensors (11, 12, 13); evaluating (22) the environmental sensor data to determine whether deceleration of the vehicle is required to avoid or mitigate a possible collision with an object, whereby a separate evaluation is performed based on the environmental sensor data of each of the several environmental sensors (11, 12, 13) and a respective sensor-specific decision is made regarding the necessity of the deceleration;Generating control signals to trigger (23) a deceleration intervention when a predetermined first number of sensor-specific decisions indicate that deceleration is required, wherein the deceleration intervention is carried out according to the control signals such that any deceleration of the vehicle thereby caused is subject to a first predetermined deceleration limit; and when, during the deceleration process, one or more further sensor-specific decisions are added to the first number of sensor-specific decisions indicating that deceleration is required, generating control signals to continue (24) the deceleration intervention while adhering to a second predetermined deceleration limit that is less restrictive than the first predetermined deceleration limit.

Description

[0001] The invention relates to a method and a processing device for controlling an automated driving function of a vehicle. Furthermore, the invention relates to a computer program with program code for carrying out such a method when the computer program runs on a software-controlled processing device, as well as a computer-readable (storage) medium containing such a computer program.

[0002] Motor vehicles are frequently equipped with automated driving functions that enable automated longitudinal and / or lateral control. For example, some modern vehicles already have automated driving functions according to SAE Level 3, which take over longitudinal and lateral control in certain driving situations without the driver having to constantly monitor the system. Such advanced automated driving functions can include automated longitudinal control as one of their sub-functions.

[0003] Such automated driving functions can also be designed to handle safety-relevant scenarios, such as the sudden braking of the vehicle in front. In conventional driver assistance systems or automated driving functions, a system reaction, for example, automatic braking in response to an object, is based on an environmental model determined from data collected by multiple sensors. Typically, the various environmental sensors are fused before a system decision is made. For example, several object lists, each calculated from the environmental sensor data of one of the sensors, can be merged into a single, unified object list. This object fusion process allows the respective strengths and weaknesses of the individual environmental sensors to be utilized and taken into account. The automated driving function then operates solely with this fused object list, which is as reliable as possible.Specifically, based on the merged object list, an object directly relevant to vehicle control is selected, and a decision is made with regard to the relevant object as to whether a system reaction, such as braking, should take place.

[0004] Patent application DE 10 2021 126 820 A1 further discloses a method for controlling a driver assistance function, in which environmental sensor data from at least two environmental sensors are received and evaluated to determine whether intervention in the vehicle's steering is necessary. A separate evaluation is performed based on the environmental sensor data from each of the at least two environmental sensors, and a sensor-specific decision is made regarding intervention in the vehicle's steering for each sensor. The multiple sensor-specific decisions are then fused into an overall decision regarding intervention in the vehicle's steering, and a control signal is generated based on this overall decision to execute the intervention.

[0005] One challenge with automated driving functions of the type described above is minimizing the risk of a rear-end collision. Heavy braking, which a vehicle performs automatically to prevent actual or perceived imminent collisions to the front, carries a certain risk of collisions with traffic behind, as a following vehicle may not decelerate in time or with sufficient force.

[0006] In particular, a hard braking maneuver performed by the automated driving function can also be a false braking maneuver, such as braking onto a so-called ghost object in the environment model. Automatic braking onto ghost objects can occur because sensor information is always subject to uncertainties. For example, if the ego vehicle's environmental sensors report a small obstacle in front of it, there is a certain probability that there is no obstacle at all, but rather that the sensor measurement was distorted by unevenness in the road, lane markings, glare from signs, reflections, or other sensor effects. Even though the probability of such false-positive object detections can be greatly reduced by fusion algorithms, a small residual probability remains.A sudden braking maneuver for a ghost object comes as a complete surprise to the traffic behind and therefore potentially poses an even greater risk of a rear-end collision.

[0007] Conventional automated emergency braking (AEB) systems release the brakes once a certain speed reduction is reached, such as 60 km / h. In the case of a ghost object, the driver would typically recognize by this point that continued braking is not necessary and would not continue the deceleration process. However, this is not possible with automated driving functions of SAE Level 3 or higher, because the driver is no longer a fallback option.

[0008] In addition to the problem described above regarding potential false positive detections of a deceleration-relevant object, there is also the challenge of avoiding false negative assessments of the situation. This means preventing the vehicle from not braking automatically, or not braking strongly enough, in situations where the presence of a deceleration-relevant object actually necessitates it. Here, it is important to strike an appropriate balance between false positives and false negatives.

[0009] It is an object of the present invention to provide a method for controlling an automated driving function which addresses the above-mentioned challenges in connection with automated braking interventions.

[0010] The problem is solved by the subject matter of the independent claims. Advantageous embodiments are specified in the dependent claims.

[0011] A first aspect of the invention relates to a computer-implemented method for controlling an automated driving function of a vehicle.

[0012] The term "vehicle" is understood to mean, in particular, a land vehicle that is moved by mechanical power and is not bound to railway tracks. A vehicle in this sense can be, for example, a car, motorcycle, or tractor.

[0013] Automated driving functions can, in principle, be any driving function that is designed to automatically intervene in longitudinal control to prevent a collision with a detected object. This could include, for example, an active safety function such as an automatic emergency braking system (AEB) or a driver assistance function such as adaptive cruise control (ACC). However, it is also possible for the automated driving function to perform longitudinal control as part of a comprehensive semi-automated driving function according to SAE Level 2, which also includes lateral control components. An example of this is a Level 2 driver assistance system that combines ACC with lateral steering and lane keeping assist (LSA).

[0014] In particular, the automated driving function can be a driving function according to SAE Level 3 to 5, i.e., a driving function that is at least conditionally automated. An example of this is a highway assistant that provides conditionally automated driving according to SAE Level 3 on certain highway sections and under defined environmental conditions up to a maximum speed of, for example, 60 km / h, 90 km / h, or 130 km / h.

[0015] One step of the process is receiving environmental sensor data from at least two environmental sensors. These environmental sensors, or some of them, can be of different types, particularly sensors based on different physical operating principles, such as an ultrasonic sensor on one side and a sensor that detects electromagnetic radiation on the other. Alternatively, they can be sensors operating in different electromagnetic wavelength ranges, such as an (optical) camera on one side and a radar or lidar sensor on the other.

[0016] A further step is to evaluate the environmental sensor data to determine whether the vehicle needs to decelerate to avoid or mitigate a potential collision with an object. This means evaluating whether, within the framework of the automated driving function, braking should be initiated in response to an object detected in the vehicle's surroundings, based on the vehicle environment as described by the respective sensor data.

[0017] The object can be a moving object, such as a pedestrian, a vehicle traveling ahead or in the opposite direction, or a bicycle. However, the object can also be a stationary obstacle, such as a parked vehicle, a car tire lying on the road, or a closed barrier at a level crossing.

[0018] The deceleration can be achieved, for example, by mechanical braking using a conventional braking system and / or by recuperating electrical energy (in the case of an electric motor drive of the vehicle).

[0019] During the evaluation step, a separate evaluation is performed based on the environmental sensor data from each of the at least two environmental sensors, and a sensor-specific decision is made regarding the necessity of the delay. In other words, several separate logical channels are provided at the data processing level for evaluating the (respective) environmental sensor data and making decisions regarding vehicle longitudinal control based on this evaluation. These logical channels can also be referred to as individual sensor channels.

[0020] In particular, with regard to the evaluation of environmental sensor data and the decision-making process concerning intervention in vehicle control based on several sensor-specific decisions, the method proposed here can rely in whole or in part on the method presented in the aforementioned DE 10 2021 126 820 A1. The content of that patent application is hereby incorporated in its entirety into the present patent application.

[0021] In a further step, control signals are generated to trigger a deceleration intervention (i.e., a decelerating intervention in the vehicle guidance system) when a predetermined initial set of sensor-specific decisions indicates that deceleration is required. In this case, a brake trigger, for example, can be activated at the data processing level. The control signal can then be output directly or indirectly (i.e., after further data processing) to a longitudinal control actuator, such as the vehicle's braking system, to implement the deceleration.

[0022] The procedure stipulates that the deceleration intervention is executed according to the generated control signals in such a way that the resulting deceleration of the vehicle complies with a first predetermined deceleration limit. This deceleration limit can, in particular, refer to a maximum permissible deceleration, i.e., a predetermined maximum deceleration value, which is not intended to be exceeded during the deceleration process. The predetermined maximum deceleration value can vary over time during the deceleration process, for example, depending on the time elapsed since the start of the deceleration or on the reduction in speed since the start of the deceleration.

[0023] Furthermore, it is provided that control signals for continuing the delay intervention are generated while adhering to a second predetermined delay limit that is less restrictive than the first predetermined delay limit, if during the delay process (in particular while the object continues to be recognized as relevant for delay by the previously decisive single sensor channel(s)) one or more further sensor-specific decisions are added to the first number of sensor-specific decisions indicating that the delay is necessary, so that as a result a second number of sensor-specific decisions indicate the necessity of a (continued) delay, the second number being greater than the first number.

[0024] A second aspect of the invention is a (data) processing device configured to carry out a computer-implemented method according to the first aspect of the invention. Accordingly, embodiments of the method according to the invention can correspond to embodiments of the processing device according to the invention described in this document, and vice versa.

[0025] The processing unit may comprise one or more processors or microcontrollers configured by a suitable computer program to execute the computer-implemented method. According to some embodiments, it may also be a spatially distributed processing unit (e.g., distributed across several spaced-apart processors or microcontrollers).

[0026] According to one embodiment of the first and / or second aspect of the invention, the second delay limitation can be less restrictive than the first delay limitation in that a larger maximum delay value is provided for at least one phase of the delay process than for the first delay limitation. Thus, both the first and second delay limitations can stipulate that a respective predetermined maximum delay value is not exceeded, with the maximum delay value according to the second delay limitation being larger than the maximum delay value according to the first delay limitation.

[0027] The maximum deceleration value according to the first deceleration limit can, for example, be in the range between 6 m / s. 2 and 8 m / s 2 , such as at 7 m / s 2. In this case, the maximum deceleration value according to the second deceleration limit can be, for example, greater than 8 m / s. 2 be and generally lie within a range that may extend to the maximum physically possible delay.

[0028] Regarding the quantitative deceleration figures presented in this document, it should be noted that positive deceleration values ​​correspond to negative acceleration values. In other words, for example, the statement that a maximum deceleration value of 7 m / s² 2 is not exceeded, which is equivalent to the fact that the vehicle's acceleration is -7 m / s² 2 will not fall below a certain threshold.

[0029] In accordance with the embodiment described above, several braking stages can therefore be maintained. As long as only a smaller initial number of available sensors (e.g., two out of three sensors, i.e., "two out of three," abbreviated 2oo3) agree that deceleration is required, a strong, but not yet the maximum possible, deceleration is released. Only when a (larger) second number of sensors agree (e.g., three out of three sensors, i.e., "three out of three," abbreviated 3oo3) is the full possible deceleration, up to the physical limits, released.

[0030] Alternatively or additionally, the second delay constraint can be less restrictive than the first delay constraint by having a less limiting effect over the course of the delay process, for example by allowing a comparatively large delay over a longer section of the entire delay process than the first delay constraint.

[0031] According to one embodiment, the first deceleration limit stipulates that a first predetermined maximum deceleration value is not exceeded until a predetermined reduction in speed, which may be at least 60 km / h, has occurred, and that subsequently a second predetermined maximum deceleration value, which is smaller than the first maximum deceleration value, is not exceeded.

[0032] For example, if, in the case of 2oo3 agreement between sensor-specific braking decisions, braking is initiated with strong, but not maximum, deceleration, it can be stipulated that this strong deceleration is maintained only for a predetermined duration or until a predetermined speed reduction occurs. For example, it can be stipulated that a maximum speed of 60 km / h can be maintained with a deceleration of up to 7 m / s². 2 (First maximum deceleration value) may be reduced. Once this 60 km / h speed reduction is reached and braking is still required according to the current environmental model, the brakes will not be fully released, but the maximum permissible deceleration will be reduced to a second maximum deceleration value, which is less than 7 m / s². 2The deceleration has been reduced. This reduction can be achieved using a continuous function or in several discrete steps. For example, the deceleration for the next 30 km / h speed reduction can be set to 5 m / s². 2 and for the remaining speed reduction to 4 m / s 2 be restricted.

[0033] One advantage of the embodiment described above is that the vehicle's braking trajectory becomes more manageable for following traffic. If the vehicle initially brakes more strongly and the deceleration subsequently decreases, a strong initial trigger is created for a following vehicle to also brake. In conjunction with the reduced deceleration later on, this lowers the risk of a rear-end collision.

[0034] According to a further development of the embodiment described above, in which the first deceleration limitation is characterized by a first and a second predetermined maximum deceleration value, the second deceleration limitation provides that, after the predetermined reduction in speed has taken place, a third predetermined maximum deceleration value, which is greater than the second maximum deceleration value, is not exceeded during the further deceleration process.

[0035] The third maximum deceleration value can, for example, be the same as the first maximum deceleration value. In other words, a deceleration that is currently in effect at the time the predetermined speed reduction is reached can be maintained.

[0036] Alternatively, the third maximum deceleration value can be greater than the first maximum deceleration value, i.e., the deceleration can be further increased compared to the current deceleration at the time the predetermined speed reduction is reached, if necessary up to the physically maximum possible deceleration.

[0037] A third aspect of the invention is a vehicle system comprising at least two (preferably different) environmental sensors and a processing unit according to the second aspect of the invention. The processing unit is configured to receive environmental sensor data from the at least two environmental sensors and, according to the method of the first aspect of the invention, to evaluate the received environmental sensor data and, based on this, to generate control signals for controlling a deceleration intervention.

[0038] According to further training, the vehicle system also includes actuators designed to implement corresponding interventions in the vehicle's guidance system based on the control signals. These actuators can include, in particular, longitudinal guidance actuators, such as a braking system and a powertrain (e.g., an electric motor drive with recuperation capability).

[0039] According to one embodiment of the third aspect of the invention, the at least two environmental sensors comprise at least two different elements from the following list: a camera; a radar sensor; a lidar sensor; an ultrasonic sensor. The camera can, in particular, be an optical camera, such as an optical front camera of the vehicle. However, the use of an infrared camera is also conceivable. The radar sensor can, for example, be a long-range radar (FRR) or a short-range radar (SRR).

[0040] A fourth aspect of the invention is a computer program comprising instructions that, when executed by a processing unit, cause it to execute a method according to the first aspect of the invention. The computer program can also be divided into several separate subprograms, each of which can be executed on different, possibly spatially separated, processing units (such as several separate processors).

[0041] For example, a vehicle system according to the third aspect of the invention may comprise one or more processing devices on which a computer program according to the fourth aspect of the invention can be executed.

[0042] A fifth aspect of the invention is a computer-readable storage medium comprising instructions which, when executed by a (possibly distributed) processing unit, cause it to execute a method according to the first 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.

[0043] A sixth aspect of the invention is a vehicle with a vehicle system according to the third aspect of the invention.

[0044] In accordance with the aspects and embodiments described above, the invention is based on the consideration that a potentially deceleration-relevant object can be detected more reliably by sensors the closer the vehicle gets to the object. This means that the probability of false negatives (but also false positives) decreases with the distance between the vehicle and the object. The method proposed here utilizes this effect to continue the deceleration process with the less restrictive second deceleration limit if a further sensor-specific decision for (continued) deceleration occurs during the braking process. This can mean, for example, that the current deceleration is maintained (if a reduction in deceleration would be intended according to the first deceleration limit to mitigate a false-positive risk) or that the current deceleration is even increased.U. up to the maximum physically possible delay. In the case of a true positive, the additional sensor would normally detect the object relevant to the delay if a certain distance is not reached.

[0045] The invention will now be explained in more detail with reference to the attached drawings. Fig. Figure 1 illustrates a vehicle system in an exemplary and schematic way. Fig. Figure 2 schematically illustrates a procedure for controlling an automated driving function. Fig. 3A illustrates, by way of example and schematically, a first delay limit that applies as long as two out of three (“2oo3”) sensors detect a delay-relevant object. Fig. 3B illustrates, by way of example and schematically, a less restrictive second delay limit, which applies as soon as three out of three (“3oo3”) sensors detect a delay-relevant object. Fig. 3C illustrates, by way of example and schematically, an alternative second delay limit that applies as soon as three out of three (“3oo3”) sensors detect a delay-relevant object.

[0046] Fig. Figure 1 illustrates, by way of example and schematically, a vehicle system 1 comprising three environmental sensors 11, 12, 13. The environmental sensors 11, 12, 13 are preferably of different types, in particular based on different physical operating principles or operating in different electromagnetic wavelength ranges, such as an (optical) camera 11, a radar sensor 12, and an ultrasonic sensor 13.

[0047] Furthermore, the vehicle system 1 includes a processing unit 10, which is used to perform a Fig. The procedure illustrated in section 2 as a block diagram is set up to control an automated driving function of a vehicle. Controlling the automated driving function can, in particular, relate to controlling automatic deceleration of the vehicle to avoid or mitigate a possible collision with an object.

[0048] The vehicle system 1 can further comprise a longitudinal guidance actuator 13 which is configured to implement a corresponding delay intervention in the vehicle guidance.

[0049] In step 21 of the procedure 2, the processing unit receives environmental sensor data from at least two of the environmental sensors 11, 12, 13.

[0050] In a further step 22, the processing unit 10 evaluates the environmental sensor data to determine whether deceleration of the vehicle is necessary to avoid or mitigate a potential collision with an object in the vehicle's vicinity. A separate evaluation is performed based on the environmental sensor data from each of the environmental sensors 11, 12, 13 from which environmental sensor data was received, and a sensor-specific decision is made regarding the necessity of deceleration.

[0051] In a further step 23, the processing unit 10 generates control signals to trigger a deceleration intervention when a predetermined first number of sensor-specific decisions indicate that deceleration is required. The control signals can then be output to the longitudinal control actuator 13 to implement the deceleration intervention. The deceleration intervention is controlled by the control signals in such a way that the resulting deceleration of the vehicle complies with a first predetermined deceleration limit.

[0052] It can happen that during the deceleration process (especially while the object is still recognized as relevant for deceleration by the previously decisive individual sensor channel(s)), one or more further sensor-specific decisions are added to the first set of sensor-specific decisions, which also confirm that a deceleration is necessary. In this scenario, a second set of sensor-specific decisions now indicates the need for a (continued) deceleration, with the second set being larger than the first. According to procedure 2, in this case, control signals for continuing the deceleration intervention 24 are generated and output to the longitudinal control actuator 13, adhering to a second predetermined deceleration limit that is less restrictive than the first.

[0053] Fig. Figure 3A schematically illustrates a first delay limitation which applies according to an embodiment of method 2 as long as two of three (“2oo3”) of the environment sensors 11, 12, 13 detect a delay-relevant object, such that two of the sensor-specific decisions support a delay intervention.

[0054] In the diagram shown, the horizontal axis indicates the speed reduction in km / h since the start of the deceleration process. The vertical axis indicates a maximum deceleration value in m / s², as defined by the first deceleration limit. 2 to.

[0055] In the example shown, the vehicle decelerates from an initial speed of 130 km / h to a standstill.

[0056] The first deceleration limit stipulates a maximum speed of 60 km / h with a deceleration of up to 7 m / s². 2(First maximum deceleration value) may be reduced. Once this 60 km / h speed reduction is reached and braking is still required according to the current environmental model, the brakes will not be fully released, but the maximum permissible deceleration will be reduced to a second maximum deceleration value, which is less than 7 m / s². 2 is lowered.

[0057] In this embodiment, the reduction of the applicable maximum deceleration value, depending on the speed already reduced, occurs in two stages: The deceleration for the next 30 km / h speed reduction, i.e., up to a total speed reduction of 90 km / h, is set to 5 m / s². 2 (second maximum deceleration value) and for the remaining speed reduction until standstill, i.e., down to a total speed reduction of 130 km / h, to 4 m / s 2 limited.

[0058] Fig. 3B and Fig. Figures 3C schematically illustrate a less restrictive second delay limit, which applies according to two alternative implementation variants of procedure 2 as soon as a further sensor-specific decision is added during the delay process, so that as a result three out of three (“3oo3”) of the environment sensors 11, 12, 13 speak in favor of a (continued) delay.

[0059] In both alternatives, the second deceleration limit stipulates that after the predetermined speed reduction of 60 km / h, a third predetermined maximum deceleration value, which is greater than the second maximum deceleration value (5 m / s²), is not exceeded during the further deceleration process. 2 ) is.

[0060] At the in Fig. In the illustrated variant 3B, the third maximum deceleration value is equal to the first maximum deceleration value and is therefore also 7 m / s. 2Accordingly, a deceleration of up to 7 m / s² is possible at the time the predetermined speed reduction of 60 km / h is reached. 2 be maintained.

[0061] In contrast, the third maximum delay value is in Fig. 3C illustrated version variant 8 m / s 2 and is therefore greater than the first maximum deceleration value (7 m / s²). 2 The deceleration can therefore be even greater (up to a maximum of 8 m / s²) compared to the deceleration at the time the predetermined speed reduction of 60 km / h is reached. 2 ) are reinforced.

[0062] For example, if, up to a speed reduction of 70 km / h, initially only 2 out of 3 sensors (2003) indicate the need for braking, and then another sensor also confirms this (3003), the first 60 km / h are calculated according to the... Fig. 3A illustrated the first deceleration limit with a deceleration of up to 7 m / s. 2 reduced, a further 10 km / h will be (also according to the first deceleration limit) at up to 5 m / s 2 reduced speed and the remaining reduction in speed until standstill can either be achieved again at up to 7 m / s 2 This will occur if the in Fig. The second deceleration limit shown in 3B is provided for, or with up to 8 m / s 2 , if the in Fig. The second delay limitation shown in 3C is provided for. QUOTES INCLUDED IN THE DESCRIPTION

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

[0000] DE 10 2021 126 820 A1 [0004, 0020]

Claims

[1] Computer-implemented method (2) for controlling an automated driving function of a vehicle, comprising the steps: - Receiving (21) environmental sensor data from multiple environmental sensors (11, 12, 13); - Evaluating (22) the environmental sensor data to determine whether a deceleration of the vehicle is required to avoid or mitigate a possible collision with an object, whereby a separate evaluation is carried out based on the environmental sensor data of each of the multiple environmental sensors (11, 12, 13) and a respective sensor-specific decision is made on the necessity of the deceleration; - Generating control signals to trigger (23) a deceleration intervention when a predetermined first number of sensor-specific decisions indicate that deceleration is required, wherein the deceleration intervention is carried out according to the control signals such that any deceleration of the vehicle thereby caused is subject to a first predetermined deceleration limit; and - if, during the delay process, one or more further sensor-specific decisions are added to the first number of sensor-specific decisions, indicating that the delay is necessary, generating control signals to continue (24) the delay intervention while adhering to a second predetermined delay limit that is less restrictive than the first predetermined delay limit. [2] Method (2) according to claim 1, wherein the multiple environmental sensors (11, 12, 13) comprise at least two different elements from the following list: - a camera; - a radar sensor; - a lidar sensor; - an ultrasonic sensor. [3] Processing device (10), wherein the processing device (10) is configured to carry out a method (2) according to claim 1 or 2. [4] Processing device (10) according to claim 3, wherein the first delay limit and the second delay limit provide that a respective predetermined maximum delay value is not exceeded and wherein the maximum delay value according to the second delay limit is greater than the maximum delay value according to the first delay limit. [5] Processing device (10) according to claim 3 or 4, wherein the first delay limitation provides that • a first predetermined maximum deceleration value is not exceeded until a predetermined speed reduction has occurred; and • subsequently, a second predetermined maximum delay value is not exceeded, which is smaller than the first maximum delay value. [6] Processing device (10) according to claim 5, wherein the second deceleration limit provides that after the predetermined speed reduction has taken place, a third predetermined maximum deceleration value is not exceeded during the further deceleration process, which is greater than the second maximum deceleration value. [7] Processing device (10) according to claim 6, wherein the third maximum delay value is equal to the first maximum delay value. [8] Processing device (10) according to claim 6, wherein the third maximum delay value is greater than the first maximum delay value [9] Computer program comprising instructions which, when the computer program is executed by a processing device (10), cause it to execute a method (2) according to claim 1 or 2. [10] Computer-readable storage medium comprising instructions which, when executed by a processing device (10), cause it to execute a method (2) according to claim 1 or 2.

Images