Method for operating a vehicle's collision avoidance system, collision avoidance system and vehicle

Abstract

Method for operating a collision avoidance system (3) of a vehicle (1), wherein, depending on a collision-critical situation, an intervention strategy for reducing the severity of a detected collision imminent between the vehicle (1) and a collision object is determined, and depending on a predicted collision angle (α) between a motion vector (B1) of the vehicle (1) and another motion vector (B2) of the collision object at a predicted collision time, a distinction is made between an intervention strategy for accident reduction, an intervention strategy for the time of collision, and an intervention strategy for rear-end collision. wherein, after determining the decision strategy by means of the collision avoidance system (3), an emergency braking function or an emergency evasive maneuver function is activated depending on a determined overlap area between the vehicle (1) and the collision object at the predicted collision time, and wherein a decision to perform automatic emergency braking or automatic emergency evasive maneuvers is made depending on a predetermined threshold value with respect to the critical overlap area between the vehicle (1) and the collision object at the predicted collision time. characterized in that a critical overlap area is determined for each intervention strategy, from which point onwards, despite initiated emergency braking, the collision between the vehicle (1) and the collision object is unavoidable and that if it is determined that the threshold value with respect to the critical overlap area per intervention strategy is undershot, the emergency avoidance function of the vehicle (1) is activated.

Description

[0001] The invention relates to a method for operating a collision avoidance system of a vehicle, wherein, depending on a collision-critical situation, an intervention strategy for reducing the severity of a detected collision imminent between the vehicle and a collision object is determined, and depending on a predicted collision angle between a motion vector of the vehicle and a motion vector of the collision object at a predicted collision time, a distinction is made between an intervention strategy for accident mitigation, an intervention strategy for the time of collision, and an intervention strategy for rear-end collision prevention. The invention further relates to a collision avoidance system and a vehicle with such a collision avoidance system.

[0002] German patent DE 10 2021 002 167 A1 discloses a method for operating a vehicle's collision avoidance system. This method determines an intervention strategy to reduce the severity of a detected, imminent collision between the vehicle and an object, depending on a collision-critical situation. Furthermore, the intervention strategy is determined and applied based on the predictable influence of the object on the existing collision-critical situation, the corresponding effect of the intervention strategy on the collision-critical situation, and the dynamics of the collision-critical situation.For this purpose, depending on a predicted collision angle between a motion vector of the vehicle and a motion vector of the collision object at a predicted collision time, a distinction is made between an intervention strategy for accident reduction, an intervention strategy for the time of collision and an intervention strategy for rear-end collision.

[0003] EP 2 749 468 B1 describes how, based on information about the vehicle itself and the distance and angle of an object to be avoided, as detected by an object recognition unit, environmental information, the position, and the magnitude of the object's movement can be calculated. Based on the accuracy of the calculated position and magnitude of the object's movement, a driving strategy for avoiding the object is determined, and autonomous driving control is executed.

[0004] The invention is based on the objective of providing a novel method for operating a collision avoidance system of a vehicle, a collision avoidance system and a vehicle with such a collision avoidance system.

[0005] The problem is solved according to the invention by a method which has the features specified in claim 1, by a collision avoidance system which has the features specified in claim 4, and by a vehicle which has the features specified in claim 5.

[0006] Advantageous embodiments of the invention are the subject of the dependent claims.

[0007] A method for operating a vehicle's collision avoidance system provides that, depending on a collision-critical situation, an intervention strategy for reducing the severity of a detected, imminent collision between the vehicle and a collision object is determined. Depending on a predicted collision angle between a motion vector of the vehicle and another motion vector of the collision object at a predicted collision time, a distinction is made between an intervention strategy for accident mitigation, an intervention strategy for the time of collision, and an intervention strategy for a rear-end collision. According to the invention, after the decision strategy has been determined by the collision avoidance system, an emergency braking function or an emergency evasive maneuver function is activated depending on a determined overlap area between the vehicle and the collision object at the predicted collision time.

[0008] The procedure represents an extension for reducing the severity of a collision, whereby the procedure is enhanced by an emergency evasive maneuver function, in particular to avoid a collision between the vehicle and the object being struck. Specifically, the emergency evasive maneuver function serves to increase the potential for avoiding collisions that were previously systematically unavoidable.

[0009] The emergency evasive maneuver function can be designed in such a way that, in addition to an emergency braking intervention by the collision avoidance system, a significantly higher effectiveness of the collision avoidance system can be achieved. Furthermore, vehicle occupants are more likely to accept an incorrectly initiated emergency evasive maneuver than an incorrectly initiated emergency braking intervention.

[0010] In one embodiment according to the invention, a decision is made regarding automatic emergency braking or automatic emergency evasive action depending on a predetermined threshold value with respect to the critical overlap area between the vehicle and the collision object at the predicted time of collision. This means that, depending on the predetermined threshold value, it is specified whether an automatic emergency braking intervention or an automatic emergency evasive action is initiated by the collision avoidance system to reduce the severity of the collision.

[0011] In one embodiment of the method according to the invention, a critical overlap area is determined for each intervention strategy, beyond which a collision between the vehicle and the collision object is unavoidable despite initiated emergency braking. Such a criterion is then used to unlock, and in particular activate, the emergency evasive function of the collision avoidance system.

[0012] In one embodiment according to the invention, if it is determined that the threshold value with respect to the critical overlap area for each intervention strategy has been undershot, the vehicle's emergency evasive steering function is activated. This initiates an automatic steering intervention, particularly to avoid the impending collision.

[0013] In a further embodiment of the method according to the invention, if it is determined that the overlap area exceeds the threshold value with respect to the critical overlap area at the predicted collision time, an emergency evasive maneuver is suppressed and the emergency braking function is activated. This threshold value thus represents a decision point at which a decision is made as to whether the emergency braking function or the emergency evasive maneuver function is applied to reduce the severity of the collision.

[0014] In one version, a warning message is issued visually and / or audibly and / or haptically in the vehicle prior to the respective intervention strategy, so that occupants of the vehicle, which is driving in automated mode, for example, are informed about the impending intervention of the collision avoidance system.

[0015] Furthermore, the invention relates to a collision avoidance system for carrying out the method, wherein, depending on a collision-critical situation, an intervention strategy for reducing the severity of a detected collision imminent between the vehicle and a collision object is determined, and depending on a predicted collision angle between a motion vector of the vehicle and a motion vector of the collision object at a predicted collision time, a distinction is made between an intervention strategy for accident reduction, an intervention strategy for the time of collision, and an intervention strategy for rear-end collision.According to the invention, a control unit of the collision avoidance system is designed to activate an emergency braking function or an emergency evasive steering function at the predicted time of collision, after determining the decision strategy depending on a determined overlap area between the vehicle and the collision object.

[0016] Using such a collision avoidance system, it is possible to initiate either the emergency braking function or the emergency evasive maneuver function, depending on the overlap area, in order to reduce the severity of the collision for the vehicle and the collision object, in particular another vehicle.

[0017] Furthermore, the invention relates to a vehicle with a collision avoidance system which is designed to activate an emergency braking function or an emergency evasive steering function at the predicted time of collision after determining the decision strategy depending on a determined overlap area between the vehicle and the collision object.

[0018] Both functions of the collision avoidance system serve to reduce the severity of a detected collision that is imminent between the vehicle and the collision object.

[0019] In particular, the emergency evasive maneuver function of the collision avoidance system is activated when a potential collision object is parked to the side of the vehicle's lane in urban traffic and protrudes into the vehicle's lane with a relatively small overlap. In such a situation, the collision avoidance system would typically initiate an emergency braking intervention, especially much earlier than a driver would be able to avoid the potential collision object with a relatively small steering movement.

[0020] Exemplary embodiments of the invention are explained in more detail below with reference to drawings.

[0021] This shows: Fig. 1 schematically two vehicles moving towards each other with their respective motion vectors and a collision angle of 180°, Fig. 2 schematically depicts a collision-critical situation with two vehicles driving towards each other and Fig. 3. Schematic overview of different collision angles.

[0022] Corresponding parts are marked with the same reference symbols in all figures.

[0023] The Fig. 1 and Fig. Figure 2 shows two vehicles, 1 and 2, in a collision-critical situation, where at least one vehicle, 1, has a collision avoidance system, 3. Additionally, a motion vector, B1, B2, is shown for each of the respective vehicles, 1 and 2. One motion vector, B1, is assigned to vehicle 1, and another motion vector, B2, is assigned to another vehicle, 2, which is a potential collision object. The two motion vectors, B1 and B2, result from the respective direction and velocity of the two vehicles, 1 and 2, at a predicted collision time.

[0024] A collision avoidance system 3 is particularly capable of managing complex accident scenarios, with the goal of achieving 360° collision avoidance with respect to vehicle 1. However, the collision avoidance system 3 must not interfere with any functions of vehicle 1, such as steering and / or braking, nor may the driver of vehicle 1 be patronized by the collision avoidance system 3. For these reasons, it is not possible to apply the same intervention strategy in every collision-critical situation.

[0025] A method for operating such a collision avoidance system 3 is known from the prior art, wherein an intervention strategy corresponding to the collision-critical situation is determined and applied.

[0026] When designing the collision avoidance system 3 with regard to the various collision-critical situations, it is relatively important to determine to what extent a potential collision object, i.e., the other vehicle 2, can have a positive or negative influence on the collision-critical situation and how a planned intervention strategy would change the collision-critical situation. Furthermore, it is relatively important to understand the dynamics of the collision-critical situation. Here, the question arises whether it can be assumed that the existing collision-critical situation will persist for a certain period of time, or whether it must be assumed that the current collision-critical situation can change for the better or for the worse at any time.

[0027] This determines in which collision-critical situations the collision avoidance system 3 reacts relatively early, resulting in a high collision avoidance potential. The more likely and unpredictable external influences are, the more conservatively the collision avoidance system 3 must be designed.

[0028] The procedure is described using a collision avoidance system 3, which includes a lane keeping system and an emergency braking system.

[0029] The lane keeping system is specifically designed to defuse a collision-critical situation by means of a steering intervention. For example, if vehicle 1 is traveling towards an adjacent lane in which another vehicle 2 is located, but which is not visible to vehicle 1 due to the blind spot, a steering intervention is automatically initiated to move vehicle 1 back into its lane.

[0030] The lane keeping system also automatically initiates a steering intervention if vehicle 1 is moving towards an oncoming lane and another vehicle 2 is approaching vehicle 1 in the oncoming lane.

[0031] Fundamentally, an automatic steering intervention by the lane keeping system always involves an initiated lane change of vehicle 1. Therefore, the lane keeping system is not notified when a potential collision object is located in the lane of vehicle 1.

[0032] Such collision-critical situations are addressed by the emergency braking system, although its effectiveness is limited by a so-called situational dilemma. These are generally situations in which braking intervention is initiated earlier than an average driver would react to the situation.

[0033] An example scenario involves another vehicle, 2, parked to the side of vehicle 1's lane, with a relatively small overlap into vehicle 1's lane. In such a situation, the driver of vehicle 1 would initiate a steering maneuver to avoid the other vehicle significantly later than the emergency braking system would initiate emergency braking to prevent a collision between vehicle 1 and vehicle 2.

[0034] The following describes a procedure for operating the collision avoidance system 3, which provides for an automatic and appropriate response to such situations with an emergency braking intervention or an emergency evasive maneuver.

[0035] For a collision avoidance system 3, the primary objective is to avoid posing any danger to traffic. Furthermore, the design of the collision avoidance system 3 is such that its intervention strategy improves, rather than worsens, the collision-critical situation, at least for vehicle 1. Given this premise, the goal is to avoid a collision wherever possible, or at least to reduce its severity. Based on this approach, three intervention strategies for the collision avoidance system 3 can be derived to achieve these requirements. These three strategies are accident mitigation, collision timing, and rear-end collision.

[0036] Regarding the intervention strategy of accident mitigation, only a reduction in the severity of the collision is possible. Prior to the intervention strategy, a warning can be issued visually, audibly, and / or haptically in vehicle 1, so that the driver of vehicle 1 is warned of an impending collision between vehicle 1 and the other vehicle 2 as the other party involved in the accident, i.e., the object of the collision.

[0037] The intervention point of the collision avoidance system 3, in particular the emergency braking system, i.e. a braking point, is chosen so late that a collision with the other vehicle 2 is unavoidable.

[0038] The intervention strategy of accident mitigation is prescribed when the questions decisive for the design are answered with a relatively high probability of unforeseen external influences. The other vehicle 2, as the collision object, can, through its interaction, have a comparatively large influence on the collision-critical situation and its outcome. Furthermore, an intervention strategy without the interaction of the other vehicle 2 cannot prevent the collision, and there is even potential to worsen the collision-critical situation. An example of this is the one in Fig. Figure 1 shows a critical collision situation involving the two vehicles 1 and 2.

[0039] In this example, it can be assumed that both the driver of vehicle 1 and the driver of the other vehicle 2 are interacting. Furthermore, in such a collision-critical situation, there is the possibility of mitigating the impact through spontaneous evasive action. Premature braking intervention by the collision avoidance system 3 could restrict the possibility of evasive action for both vehicle 1 and the other vehicle 2, so early braking intervention by the collision avoidance system 3, especially the emergency braking system, should be avoided.

[0040] Even if vehicle 1 is brought to a standstill, the braking intervention, i.e., an intervention of the emergency braking system, has no influence on the collision object, that is, on the other vehicle 2. The collision of the two vehicles 1 and 2 occurs even without interaction from the other vehicle 2.

[0041] Due to the dynamic nature of the situation, the emergency braking system can reduce maximum impact energy through a comparatively very late braking intervention. Furthermore, this intervention strategy is limited by the requirement that the collision must occur between vehicle 1 and the other vehicle 2. This uncertainty is technically optimized by determining the overlap area between vehicle 1 and the other vehicle 2 at a predicted collision time. The larger the overlap area, the higher and more certain the collision potential.

[0042] With the collision-time intervention strategy, collision avoidance is possible. Here, too, a warning can be issued visually, audibly, and / or haptically prior to the intervention strategy. However, the warning has a comparatively low potential benefit. The braking point is chosen so late that the collision can be prevented with maximum physical deceleration of vehicle 1.

[0043] The intervention strategy "collision time" is chosen when the questions crucial for the design are answered with a high probability of unforeseen external influences. The other vehicle 2, through its interaction, can significantly influence the outcome of the collision-critical situation. Furthermore, the collision avoidance system 3's intervention strategy can prevent the collision without the interaction of the other vehicle 2, although this strategy has a low potential to worsen the collision-critical situation. An example of this is another vehicle 2 crossing the path of vehicle 1.

[0044] Here too, it can be assumed that both the driver of vehicle 1 and the driver of vehicle 2 will interact. In these cases, a relatively minor adjustment to the driving condition of at least one of the two vehicles (1, 2) is sufficient to prevent the collision. This involves at least one of the two vehicles (1, 2) reducing or increasing its speed. There is only a comparatively small window of time in which the collision is predictable.

[0045] Since the interactions between the driver of vehicle 1 and the other vehicle 2 cannot be anticipated by the emergency braking system of the collision avoidance system 3, or can only be anticipated very imprecisely, applying the intervention strategy too early would confuse the driver and, in the worst case, could worsen the critical collision situation. Therefore, in these cases, braking intervention as an intervention strategy can only be initiated at the latest possible point. The collision between vehicles 1 and 2 can nevertheless be avoided without interaction from the other vehicle 2 by the intervention strategy applied by the emergency braking system of the collision avoidance system 3.

[0046] Regarding the intervention strategy at the moment of collision, the goal of the emergency braking system is to avoid a collision by initiating a so-called targeted full braking maneuver. Here, too, the braking intervention is limited by the uncertainty as to whether a collision between vehicle 1 and the other vehicle 2 will occur.

[0047] The intervention strategy for rear-end collisions offers the greatest potential for collision avoidance through braking intervention by the emergency braking system of the collision avoidance system 3. Prior to the execution of this intervention strategy, a warning is issued visually, audibly, and / or haptically. The issuance of this warning has a relatively high potential benefit, as the timing of its issuance is chosen so early that, under optimal circumstances, the occupant of vehicle 1 can defuse the critical collision situation themselves.

[0048] In a further stage of the collision-critical situation, an emergency braking maneuver is initiated early by applying autonomous partial braking. If the autonomous partial braking is insufficient to prevent the collision of the two vehicles 1 and 2, autonomous braking with maximum deceleration is initiated.

[0049] This intervention strategy is chosen when the crucial questions for the design of the collision avoidance system 3, particularly the emergency braking system, can be answered with a low probability of unforeseen external influences. The other vehicle 2, as the collision object, can only have a relatively minor influence on the outcome of the critical collision situation through its interaction. In such a critical collision situation, intervention by the emergency braking system has a high potential to prevent the collision between the two vehicles 1 and 2. Conversely, the intervention has little to no potential to worsen the critical collision situation.

[0050] A typical example of the application of the intervention strategy "collision situation" is the approach of vehicle 1 to a preceding vehicle 2 at a relatively high relative speed.

[0051] Here, the braking intervention always contributes to defusing the collision-critical situation. It can be assumed that the other vehicle 2 will not interact to avoid the collision, as the driver of the other vehicle 2 may not be aware of the collision-critical situation. Furthermore, the physical possibilities of the other vehicle 2 to defuse the collision-critical situation by changing its driving state are rather limited.

[0052] Here too, the goal of collision avoidance system 3, and in particular the emergency braking system, is to prevent a collision between vehicle 1 and the other vehicle 2. In addition to issuing a warning early and initiating braking early, the system attempts to further maximize the collision avoidance potential. In these collision-critical situations, uncertainty increases as the overlap between vehicle 1 and the other vehicle 2 decreases. Therefore, the following also applies here: the greater the overlap between the two vehicles 1 and 2 at the predicted collision time, the higher the collision potential.

[0053] A key parameter is the collision angle α of the two motion vectors B1, B2 of vehicles 1, 2 at a predicted collision time. Fig. Figure 3 shows an overview of a plurality of collision angles α of the motion vectors B1, B2.

[0054] The collision angle α is 0° when the motion vectors B1, B2 are parallel to each other, since the two vehicles 1, 2 are traveling parallel to each other.

[0055] If the other vehicle 2 crosses the path of vehicle 1 from the left, the collision angle α is 90°, whereas it is 270° if the other vehicle 2 crosses the path of vehicle 1 from the right. If the two vehicles 1 and 2 approach each other, as in Fig. As shown in Figure 1, the collision angle α of the motion vectors B1, B2 is 180°.

[0056] By specifying value ranges with respect to the collision angle α, it is possible to uniquely assign a respective intervention strategy of the collision avoidance system 3 to a specific value range of the collision angle α determined on the basis of the motion vectors B1, B2.

[0057] For example, the collision mitigation intervention strategy is applied when the collision angle α is greater than 170° and less than 190°. The collision point intervention strategy is applied by the collision avoidance system 3 when the collision angle α is greater than 0° and less than 170°, or when the collision angle α is greater than 190° and less than 360°. In the event that the collision angle α is greater than 315° and less than 360°, or greater than 0° and less than 45°, the rear-end collision intervention strategy is selected and applied. The ranges of values ​​for the collision angle α for applying the rear-end collision intervention strategy are a subset of the ranges of values ​​for applying the collision time intervention strategy.

[0058] By relatively simple adjustment of the value ranges, the behavior of collision avoidance system 3 can also be adapted relatively quickly and easily. This allows the functionalities explained below to be implemented.

[0059] The parameterization of the collision angle α value ranges can be performed using external control unit calibration values. This allows the respective value range to be adjusted via diagnostics during the production run of vehicle 1 or at a later time. In field operation, this enables relatively quick, simple, and predictable responses to problems by redistributing the intervention strategies across the collision angle α value ranges. The rationale for this is that the collision mitigation intervention strategy and the collision timing intervention strategy have a significantly higher risk of false triggering.

[0060] The procedure described above allows for the prediction of the behavior of collision avoidance system 3, particularly the emergency braking system. Furthermore, it largely eliminates the risk of violating official expectations of functionality with regard to legal requirements and so-called rating criteria.

[0061] Furthermore, the subdivision of intervention strategies based on the respective value range of the collision angle α provides a basis for future functional enhancements, such as automated avoidance. Depending on the physical possibilities arising from these functional enhancements, the value ranges of the collision angle α are adjusted accordingly with regard to the application of the respective intervention strategy.

[0062] A key element for deciding whether the collision avoidance system 3 initiates an emergency braking intervention or an emergency evasive maneuver is the overlap area between vehicle 1 and the other vehicle 2, the potential collision object, at the predicted time of collision.

[0063] Since a system design for the emergency braking system of the collision avoidance system 3 is known, a critical overlap range can then be determined for each of the intervention strategies, from which the emergency braking system is no longer able to completely avoid a collision between vehicle 1 and the other vehicle 2 by initiating an emergency braking intervention.

[0064] This critical overlap area, in particular a threshold specified with respect to it, is used to unlock, that is, activate, an emergency avoidance function.

[0065] If the overlap area is sufficiently large at the predicted collision time, exceeding the specified threshold, the emergency avoidance intervention of the collision avoidance system 3 is suppressed, as the existing collision-critical situation is already adequately addressed via the emergency braking function.

[0066] If a collision-critical situation arises where the determined overlap area falls below the predefined threshold, an emergency evasive maneuver is automatically initiated by a control unit of the collision avoidance system 3 to prevent a collision between vehicle 1 and another vehicle 2, which, for example, is partially parked in one of vehicle 1's lanes. By means of the automatically initiated emergency evasive maneuver, vehicle 1, taking its surroundings into account, avoids the other vehicle 2 to prevent a collision between vehicle 1 and the other vehicle 2.

[0067] This emergency evasive action function can be designed in such a way that, in addition to the emergency braking intervention, a significantly higher effectiveness of the collision avoidance system 3 can be achieved. Furthermore, a comparatively minor, erroneous emergency evasive action is more readily accepted by the occupants of vehicle 1 than an erroneously initiated emergency braking intervention by the collision avoidance system 3.

Claims

[1] Method for operating a collision avoidance system (3) of a vehicle (1), wherein, depending on a collision-critical situation, an intervention strategy for reducing the severity of a detected collision imminent between the vehicle (1) and a collision object is determined, and depending on a predicted collision angle (α) between a motion vector (B1) of the vehicle (1) and another motion vector (B2) of the collision object at a predicted collision time, a distinction is made between an intervention strategy for accident reduction, an intervention strategy for the time of collision, and an intervention strategy for rear-end collision. wherein, after determining the decision strategy by means of the collision avoidance system (3), an emergency braking function or an emergency evasive maneuver function is activated depending on a determined overlap area between the vehicle (1) and the collision object at the predicted collision time, and wherein a decision to perform automatic emergency braking or automatic emergency evasive maneuvers is made depending on a predetermined threshold value with respect to the critical overlap area between the vehicle (1) and the collision object at the predicted collision time. characterized by , that a critical overlap area is determined for each intervention strategy, from which point onwards, despite initiated emergency braking, the collision between the vehicle (1) and the collision object is unavoidable and that if it is determined that the threshold value with respect to the critical overlap area per intervention strategy is undershot, the emergency avoidance function of the vehicle (1) is activated. [2] Method according to claim 1, characterized by , that if it is determined that the overlap area exceeds the threshold with respect to the critical overlap area at the predicted collision time, an emergency evasive maneuver is suppressed and the emergency braking function is activated. [3] Method according to any one of the preceding claims, characterized by , that a warning is issued visually and / or audibly and / or haptically in the vehicle (1) prior to the respective intervention strategy. [4] Collision avoidance system (3) for carrying out the method according to any one of claims 1 to 3, characterized by, that a control unit is designed to activate an emergency braking function or an emergency avoidance function at the predicted time of collision, after determining the decision strategy depending on a determined overlap area between the vehicle (1) and the collision object. [5] Vehicle with a collision avoidance system (3) according to claim 4.

Images