Method for determining a trigger criterion for braking and an emergency braking system for a vehicle

Abstract

Method for determining a trigger criterion (k_B) for outputting brake signals (S4, S5) in a vehicle (1), comprising at least the following steps: Detecting at least one object (8, 18) in the vicinity of the vehicle (1) (St1), Determine whether the vehicle (1) is on a collision course with the object (8, 18) (St2), when determining a collision course with the detected object (8, 18), checking an avoidance criterion (K_avoid) which is met if no avoidance maneuver for the vehicle (1) can be determined or is possible (St3), at least if the avoidance criterion is met, check a braking criterion (K_brake) (St8), the object (8, 18) is classified into at least a first or a second class depending on its properties (St3), where the evasive maneuver of the vehicle (1) is determined as a function of the classification of the object (8, 18) in such a way that, when classified in a first class, a first evasive trajectory (10) is determined and, when classified in a second class, a second evasive trajectory (11) different from the first evasive trajectory is determined (St4, St5, St6, St7), wherein the avoidance criterion (K_B) for outputting braking signals (S2, S4) is met when the avoidance criterion (K_avoid) and the braking criterion (K_brake) are met, characterized in that at least one extreme value (x_q1, x_q2, q(v,r)) of the lateral acceleration (q) acting on the vehicle is determined from the determined first or second avoidance trajectory (10, 11) and as an avoidance criterion (K_avoid) the at least one extreme value (x_q1, x_q2, q(v,r)) is compared with a lateral acceleration limit value (q_g), wherein the avoidance criterion (K_avoid) is met if the at least one extreme value (x_q1, x_q2, q(v,r)) is greater in magnitude than the lateral acceleration limit value (q_g).

Description

[0001] Some vehicles, especially commercial vehicles, are equipped with emergency braking systems. For this purpose, the vehicles generally have environmental sensors that specifically monitor the area in front of the vehicle and detect possible collisions with the detected objects. If a collision course is identified, an automatic emergency braking system can be triggered to prevent a collision or at least reduce the impact force.

[0002] However, in a collision course, a driver may still be able to avoid the object instead of braking. Thus, evasive maneuvers can be determined instead of or in addition to emergency braking.

[0003] DE 10 2004 056 027 A1 describes a vehicle assistance system for preventing collisions or reducing the severity of collisions. Depending on the detection, warning signals are issued to the driver and / or automatic steering and / or braking intervention can be performed. Thus, instead of braking, an automatic steering maneuver for an evasive maneuver can be performed or displayed to the driver.

[0004] DE 10 2004 028 404 A1 describes the determination of an evasive trajectory, for which the trajectory of a vehicle in front is determined. DE 10 2010 006 214 A1 describes an emergency braking assistant for automatic braking, which takes into account a driver reaction time that must be taken into account before the driver initiates a braking maneuver. This calculates the driver reaction time, a braking intervention time, and an evasive maneuver time. DE 103 369 86 A1 describes a method for avoiding collisions in a vehicle, in which information about the movement behavior of objects in the vehicle's surroundings is stored.DE 100 127 37 B4 describes a device for performing a lane change by a vehicle, in which a trajectory planning device is used to generate a transition curve signal, taking into account that a lateral acceleration specified by a driver of the motor vehicle is not exceeded. A substantially S-shaped transition trajectory is determined, in which the tangent of the transition trajectory at the end of the lane change is equal to the determined tangent of the lane. The transition trajectory is applied as a third-order polynomial to determine a sigmoid or S-shaped function of the lateral offset as a function of a longitudinal parameter along the roadway.

[0005] DE 101 543 21 B4 describes a vehicle guidance support control or regulation system in which, when an obstacle is detected on the roadway, information about the obstacle is obtained and an avoidance route is determined, using map information and determining the avoidance route as a curvature with a sinusoidal mathematical function of the position of the vehicle along the route.

[0006] DE 10 2010 023 164 A1 describes a method for warning a driver of a motor vehicle, in which possible collisions with detected objects are determined and a warning criterion is taken into account using a calculated path length to the detected object, whereby a steering angle is used. DE 10 2010 028 384 A1 describes the control of the driving stability of a vehicle and the determination of an evasive trajectory for the vehicle in critical situations. Depending on the critical situation, yaw angles are used, target steering angles are calculated, and control thresholds are reduced as preparatory measures. EP 1 057 159 B1 describes the prevention of a collision between a vehicle and an obstacle, whereby several distances to the vehicle are determined in order to take into account, on the one hand, braking with maximum deceleration and, on the other hand, steering around the obstacle; half overlaps of the vehicles can be used here.

[0007] EP 1 303 421 B1 describes an automatic braking and steering system in a vehicle in which evasive paths around an obstacle are determined and additionally taken into account if there is another obstacle in the evasive path. EP 1 223 093 B1 describes an automatic braking system with surroundings detection and the detection of lane changes and emergency braking. This determines whether the driver intends to change lanes and, if so, suppresses emergency braking. EP 1 409 311 B1 describes a method for avoiding or mitigating a collision in which objects are detected and classified into various classes, for example vehicle, motorcycle, stationary building, with the classification being carried out by means of shape detection by radar and determination of the speed of the objects. EP 1 263 634 B1 describes a driving stability control system in which a vehicle trajectory is determined and influenced by wheel interventions.Map and radar information can be used to record the road surface.

[0008] EP 1 409 310 B1 describes a method for predicting the movement trajectories of a detected object, whereby only possible trajectories of the object are determined in which the forces acting on the object or detected vehicle lie within the maximum transferable values. EP 1 926 647 B1 also describes a method for determining a movement trajectory in which the path of the detected object is extrapolated using polynomials, exponential functions, or trigonometric functions.

[0009] EP 1 926 646 B1 describes a method for carrying out an evasive maneuver using hyperbolic tangent functions and logistic functions as well as arctangent functions.

[0010] US 7 283 902 B2 describes the determination of evasive maneuvers taking into account the yaw rate.

[0011] EP 1 926 654 B1 describes further mathematical calculations.

[0012] US 8 170 739 B2 describes the determination of an evasive trajectory using mathematical calculations including yaw angle and steering angle, using fifth-order polynomials.

[0013] WO 2012 / 119 596 A1 describes the evaluation of an evasive option in highly dynamic situations by predicting hypotheses and predictively displaying evasive options.

[0014] DE 102 31 557 A1 discloses a method and device for initiating and executing a deceleration of a vehicle to avoid a collision or reduce the severity of an accident. Environmental sensors detect objects within the sensor detection range. The objects are assigned to different object classes, and based on the assignment of the detected objects to the respective object class, the objects' movement trajectories are predicted. From these predicted movement trajectories of the objects and their associated, detected object classes, a collision risk and a hazard level are determined. If predefined combinations of collision risk and hazard level are present, the vehicle's deceleration devices are activated accordingly.

[0015] However, initiating automatic steering maneuvers is generally quite complex and requires automatically controlled steering systems. Performing an emergency braking maneuver when an evasive maneuver might still have been possible can generally lead to unnecessary problems and damage, for example, from vehicles rear-ending.

[0016] The invention is based on the object of creating a method for determining a trigger criterion for braking and an emergency braking system that enable a high level of safety. This object is achieved by a method according to claim 1, an emergency braking system according to claim 17, and a vehicle according to claim 18. The subclaims describe preferred developments.

[0017] The invention is based on the idea of ​​suppressing the output of brake signals as warning signals or brake control signals of an automatic emergency braking system for as long as the driver is still able to evade the collision. This can at least largely prevent false activations.

[0018] According to the invention, detected objects are classified differently in order to determine different avoidance trajectories depending on the classification. In particular, a distinction is made between a moving object and a stationary object. Depending on this distinction or classification, different forms of avoidance trajectories or different mathematical formulas for the avoidance trajectory are applied.

[0019] The idea behind this is that it can be assumed that a moving object is traveling in a lane in front of the vehicle, in particular in the same lane as the vehicle. An evasive trajectory to avoid a collision should lead past this object and then return to roughly the same direction of travel; thus, the evasive trajectory can be determined as a lane change or driving past the moving object and then cutting in. The evasive trajectory is advantageously determined with a turning point or in an S-shape (sigmoid). It is recognized that such S-shaped evasive trajectories have a turning point in a central area and a point of extreme lateral acceleration before and behind it, namely maximum right-hand curvature and left-hand curvature. The extreme value orThe extreme value of lateral acceleration is defined as a value of lateral acceleration that represents a relative maximum value in terms of magnitude, i.e., is adjacent to lateral acceleration values ​​with lower magnitudes. These two extreme values ​​can therefore be used directly for the avoidance criterion and compared with each other and the lateral acceleration limit value.

[0020] If, however, the detected object is classified as stationary, the invention recognizes that an evasion trajectory can be applied that essentially describes cornering; thus, in particular, an evasion trajectory with a constant radius of curvature or a radius of curvature within a tolerance range can be used. The idea underlying this is that a traffic sign or, if applicable, an object of peripheral development is identified as the stationary object, and a lane can be selected that leads past the object. When using such an evasion trajectory with an essentially constant radius of curvature, an essentially constant lateral acceleration will thus have to be applied while driving on the evasion trajectory - at essentially the same driving speed.It is recognized that such a curved path leads to smaller lateral accelerations compared to the S-shaped avoidance trajectory at the same driving speed; thus, when a stationary object is detected, a smaller minimum distance can be permitted until the avoidance criterion is met and avoidance is no longer possible.

[0021] To determine an evasive trajectory involving cornering, additional map data and navigation data can be used to verify the presence of such a lane leading essentially in a circular arc. However, the invention recognizes that, in principle, such an evasive trajectory can also be applied without map data, since such map data is often inaccurate or out of date. This, in turn, is based on the idea that the invention aims to prevent the false triggering of an emergency braking maneuver and also to take into account such potentially existing evasive options.

[0022] According to a further development, several detected objects that are sufficiently close to each other can be detected or evaluated as a common object, and thus an avoidance trajectory can be placed around the several objects detected together.

[0023] According to the invention, the avoidance trajectories are advantageously determined relative to the vehicle's own projected lane (own trajectory) or the driving path in front of the vehicle. For example, when cornering, it is taken into account that an avoidance trajectory must be applied as a change compared to the projected own trajectory. For this purpose, a curve parameter along the lane can be set as a parameter instead of the linear x-direction along a straight roadway.

[0024] Furthermore, lane routing considerations can be used to take partial overlapping of vehicles into account.

[0025] The invention is explained below with reference to some embodiments in the accompanying drawings. They show: Fig. 1 a street scene in plan view with a vehicle according to the invention during the determination of an evasive maneuver around a moving object; Fig. 2 a corresponding representation when determining an evasive maneuver around a stationary object; Fig. 3 Representation of a sliding sine function for determining an avoidance trajectory according to Fig. 1, Fig. 4 a vehicle with an emergency braking system according to the invention, Fig. 5 a flowchart of a method according to the invention, Fig. 6 a representation of the minimum distances as a function of the relative speed during different maneuvers.

[0026] A vehicle 1, in particular a commercial vehicle, is driving according to Fig. 1 on a roadway 2 with three lanes 2a, 2b, 2c in a current direction of travel F, with a driving speed v. At the current time t0, vehicle 1 is traveling in a straight line, meaning its lateral acceleration q (t0) = 0.

[0027] Vehicle 1 is in Fig. 4 and has an environmental sensor system, for example, distance sensors 3 based on radar or ultrasound for detecting objects in front of the vehicle 1 and, if necessary, from the side areas outside the roadway 2. The environmental sensor system can also detect areas to the side of the vehicle 1 or behind the vehicle 1. The environmental sensor system can, for example, also have camera systems for visually detecting the roadway. The vehicle 1 further has, according to Fig. 4 a control device 4 of a braking system or an emergency braking system 22, as well as brake actuating devices 5, which are controlled by the control device 4 by means of brake control signals S2. The control device 4 receives environmental sensor signals S1 from the environmental sensor system 3. The control device 4 also receives driving dynamics status signals S3, for example via an internal vehicle data bus, which indicate the vehicle's own driving speed v, and also, for example, a longitudinal acceleration ax and / or lateral acceleration q. In this case, the control device 4 can also determine the driving speed v, for example, from wheel speed signals, and the longitudinal acceleration ax as a time derivative of the driving speed v.

[0028] According to one embodiment, the vehicle 1 has a navigation device 7 which determines the current position of the vehicle 1 in map data and for this purpose has either stored map data or is currently recording map data.

[0029] According to Fig. 1, at a time t0, another vehicle 8 is traveling as an active road user in front of vehicle 1 in the same, here middle lane 2b. Vehicle 1 initially detects the other vehicle 8 as object 8 using its environmental sensors 3 and stores the determined data of object 8 in a memory 6 connected to control device 4 or formed as part of control device 4. Control device 4 stores, in particular, driving characteristics, in particular data on driving speed v8, and possibly also longitudinal acceleration a8 and lateral acceleration q8 of object 8 over a past period from tz to t0.These data are used, on the one hand, to classify object 8 and, on the other hand, to determine whether a collision course exists; for this known determination of a collision course, further determined data can be used, in particular lateral speed dvy8, longitudinal distance dx8, lateral offset dy8; all data can be used relative to vehicle 1.

[0030] The control device 4 of the vehicle 1 thus detects the object 8, classifies it, determines an evasive trajectory, and decides, based on an evasive criterion K_avoid, whether an evasive maneuver is possible. Furthermore, based on a braking criterion K_brake, it decides whether an emergency braking maneuver is necessary. Furthermore, additional criteria may be applied. These include: The detection of object 8 represents in the flowchart of the Fig. 5 represents step St1. In the subsequent step St2, the control device 4 determines whether the host vehicle 1 is on a collision course with the detected object 8 (or whether the detected object 8 is on a collision course with the vehicle 1). For this purpose, a respective current distance dx8 to the detected object 8 is detected and the current trajectory of the vehicle 1 is determined and extrapolated based on its driving dynamics data, in particular driving speed v, longitudinal acceleration a, lateral acceleration q and / or possibly yaw rate ω. Accordingly, the current trajectory of the moving object 8 is extrapolated based on its driving dynamics data relative to the host vehicle, in particular lateral offset dy8, lateral relative speed dvy8, distance dx8, relative longitudinal speed dvx8 and relative longitudinal acceleration dax8, and then checked to see whether the trajectories intersect in the subsequent period, which is recognized as a collision course.In the case of the . Fig. 1 with one-dimensional movements of the vehicle 1 and the moving object 8, a second-order differential equation can thus be determined in time, that is, taking into account the current distance dx8, the differential speed v - v8, and the current longitudinal accelerations ax and a8.

[0031] If a collision course is absent, the system resets to before the first step St1 according to branch n. If a collision course is determined according to branch y, an avoidance criterion K_avoid is subsequently checked. When determining the avoidance criterion K_avoid, the detected object 8 is first classified in step St3. For this purpose, it is determined whether the detected object 8 did not exhibit a standstill at least temporarily during a previous detection period from tz to the current time t0, i.e., whether the driving speed v8 was above a measurement limit value v_m, which indicates a measurement inaccuracy and can, for example, be several km / h.

[0032] If v8 < v_m during the entire detection period, a detected object is recorded as a stationary object; Fig. 2 shows this case for a traffic sign 18 as a stationary object 18. Other objects that at least temporarily showed v8 > v_m are classified as moving objects or road users, as in Fig. 1 the moving object 8, which currently shows v8 > v_m. In principle, a distinction can be made between currently stationary moving objects, i.e., v8 (t0) < v_m, with v8 > v_m within tz to t0, e.g., vehicles at the end of a traffic jam or in an accident, and currently moving objects 8, which show v8 (t0) > v_m. However, such a differentiation in the classification is generally not required.

[0033] Depending on the classification in step St3, different avoidance trajectories are subsequently determined. Upon determination of a stationary object 18, an avoidance trajectory 11 for cornering without lane change, in particular with a substantially constant curve radius, for a stationary object 18 is subsequently determined according to branch o1 in step St4, as described in Fig. 2. Upon detection of an at least temporarily non-stationary object 8, i.e. a road user, an avoidance trajectory 10 for a lane change is determined according to branch o2 in step St5. Fig. 1 determined.

[0034] The determination of the avoidance trajectories in steps St 5 and St 4 serves to determine a maximum lateral acceleration q of vehicle 1. For this purpose, deliberately simplified trajectory functions are used for avoidance trajectories 10 and 11.

[0035] First, the case of detecting a moving object 8 according to Fig. 1. In step St 5, the avoidance trajectory 10 is determined such that a lane change takes place from the current lane 2b to an adjacent lane 2a or 2c. The final orientation of the vehicle 1 at the end of the avoidance trajectory 10 or at the end of the avoidance maneuver essentially corresponds to the current orientation. The orientation can be either the vector of the driving speed v, i.e. the direction of travel F, or the orientation of the longitudinal axis L. An avoidance trajectory 10 with a turning point x_w is used, i.e. a change in the direction of curvature of the avoidance trajectory 10. According to the invention, it is recognized here that the lateral acceleration q in the avoidance trajectory 10 has two extreme values, a first extreme value x_q1 before the turning point x_w, i.e. the curve movement in the first region of the trajectory z. B. to the left, and a second extreme value x_q2 after the inflection point x_w, iein the second area of ​​the trajectory with the opposite impact, e.g. to the right.

[0036] During an evasive maneuver from the center lane 2b to the adjacent left lane 2a, the steering wheel is initially turned to the left, resulting in an initial evasive trajectory 10 curve to the left. The lateral acceleration q thus initially increases from the current straight-ahead travel to the first extreme value x_q1 of the lateral acceleration q; subsequently, the lateral acceleration q decreases in magnitude until the turning point w is reached. At the turning point w there is a change in the direction of curvature, in this case from a left turn to a right turn or from a left turn to a right turn, so that - under ideal assumptions without, for example, skidding behavior - there is a zero crossing of the lateral acceleration q, i.e. q (t_w) = 0. Subsequently, the lateral acceleration q increases again in magnitude in the subsequent trajectory section with a right turn up to the second extreme value x_q2, whereupon the lateral acceleration q decreases again.The avoidance trajectory 10 in the first case of the . Fig. 1 is advantageously determined in such a way that it transitions smoothly or continuously differentiable into a straight-ahead drive on the adjacent lane 2a.

[0037] Fig. Figure 3 shows the particularly preferred sliding sine function y = x - sin (x) as the avoidance trajectory, ie, with respect to the coordinates x, y in the coordinate system, the additive superposition of a linear term and a trigonometric function. When displayed or calculated in a Cartesian coordinate system, the longitudinal direction of the lane, in Fig. 1 is thus taken as the Cartesian coordinate x, and the transverse direction y as the second coordinate. The avoidance trajectory 10 is thus described as a sliding sine function with the relationship y (x). In a simple case, for example, the sliding sine function y = x - sin (x) can be specified for this purpose, with y and x as dimensionless numbers. The sine is thus related to the angular unit radian: sin (π) = sin (3.1415...) = sin (2 π) = 0 and sin (π / 2) = 1. According to the invention, it is recognized that this achieves a function which, in the first period, i.e. from 0 to 2 π = 6.28..., has a function with a suitable shape, which has an inflection point, is twice continuously differentiable and also has smooth transitions at the start time x = 0 and the end time x = 2π.It is recognized as advantageous that the first and second derivatives each also represent trigonometric functions, in particular the second derivative itself is a pure trigonometric function; for the simple sliding sine function y = x - sin (x), this results in y' = 1 - x cos (x) and y'' = sin (x). The lateral acceleration is determined via the curve curvature of the evasive trajectory and the vehicle speed at the respective curve points along the x-axis:. ay(x)=ν(x)2∗κ(x) κ(x)=y(x)"(1+(y(x)')2)32

[0038] The extreme values ​​of the lateral acceleration are determined using the zeros of the first derivative of the lateral acceleration: ay'(x)=0⇒xay1;xay2⇒extreme values:ay1=ay(xay1);ay2=ay(xay2)

[0039] The x-axis represents the movement in the longitudinal direction or direction of travel, the y-axis in Fig. 3 the movement in the transverse direction. The function is subsequently stretched or extended accordingly. Thus, a distance L is used in the longitudinal direction, as well as a transverse extension dy of the transverse offset to be overcome. For the evasive maneuver, the period 2π of the sine is scaled to the distance L, and the evasive trajectory 10 is stretched in the transverse direction such that the lateral offset dy (y(L) = dy) is reached after the distance L.

[0040] The following equation results: y=dyLx−dy2πsin(2πLx)

[0041] The distance L to the object 8 can be calculated, for example, from the sum of the current distance dx at time t and the distance that the object travels in the time that the vehicle needs to reach the object.

[0042] The necessary lateral offset dy can, for example, be composed of the sum of half the width of the own vehicle, half the width of object 8, the transverse offset between the own vehicle and the object at the time the own vehicle reaches the object and any safety distance.

[0043] In step St6, the magnitudes of the maximum acceleration values ​​q1 and q2 are subsequently compared with a lateral acceleration limit value q_g, and it is determined whether the permissible lateral acceleration limit value q_g is exceeded. To do this, the maximum values ​​of q1 and q2 are first determined and then compared with q_g.

[0044] The avoidance criterion for subsequent initiation of emergency braking is reached when the lateral acceleration limit value q_g is exceeded and thus no further avoidance maneuver around a moving object 8 is possible. Thus, in step St6, if the avoidance criterion K_avoid is not met, whereby avoidance is still possible, the system is reset according to branch n, and if the avoidance criterion K_avoid is met, a braking criterion is subsequently examined according to branch y.

[0045] If it is determined in step St3 that a stationary object 18 has been detected, then instead of an avoidance trajectory with a turning point, a Fig. 2 shown avoidance trajectory 11 with essentially the same curve radius R3 is determined in such a way that here too the avoidance trajectory 11 passes the detected stationary object 18. In particular, in this case too the Fig. 2, additional map data can be used to determine whether the lane of the determined avoidance trajectory 11 lies on a permissible lane 2d, for example an exit or junction. The determination according to Fig. 2 can, however, also be carried out without such map data being available, in particular if an exact positioning is not certain or not possible. Thus, according to Fig. 2 the widening trajectory 11 is determined as a right-hand curve with a constant radius R3, so that the lateral acceleration q occurring when driving on the avoidance trajectory 11 is given as a function of the distance s driven on the avoidance trajectory as q = v(s) 2 / R3.

[0046] The lateral acceleration q is therefore constant on the avoidance trajectory 11 at a constant driving speed v; if v changes, for example due to a braking process, it changes accordingly. Fig. 2 determined avoidance trajectory 11 with constant curve radius R3 allows a smaller minimum distance L2 than the minimum distance L1 in the case of Fig. 1; at the same current driving speed v in the cases of Fig. 1 and Fig. 2 will be L2 < L1 when the lateral acceleration limit q_g is reached. This means that the avoidance criterion K_avoid is Fig. 2 is not met so quickly; small minimum distances L2 can be reached without the avoidance criterion being met. Thus, in a step S7 following step St4, the current lateral acceleration q (v, R3) is compared with the permissible limit lateral acceleration q_g; if this is undershot, the process is reset according to branch n, and if it is exceeded according to branch y, further criteria are subsequently checked in step St8; otherwise, the process is reset according to branch n.

[0047] In step St8, the braking criterion K_brake is first checked to determine whether an emergency braking procedure should be initiated; in particular, an emergency braking procedure with maximum braking force a_br can be initiated, ie, for example, a path equation in second order of time is applied with L1 or L2, v and a, e.g. as L1< (a_br*t 2 ) / 2+vt.

[0048] Furthermore, a time delay can be taken into account by an actuation time t_act of the brake actuators and, if necessary, also a reaction time t_react of the driver.

[0049] Thus, in step St8, it is checked whether emergency braking is possible in front of the detected object 8 or 18. For this purpose, automatic emergency braking and emergency braking initiated by the driver after a warning can be examined. In a first step, it can therefore be checked whether, taking into account the human reaction time t_react, the driver cannot initiate emergency braking himself, i.e. if a brake warning signal S4 is output to the driver by the control device 4, for example to a display device 20 in the cockpit area. Furthermore, it can be checked whether automatic emergency braking can be initiated by a brake control signal S2.

[0050] If K_avoid and K_brake are fulfilled, a trigger criterion K_B for braking the vehicle is fulfilled. Thus, in the simplest case, K_B is given by K_avoid and K_brake, ie K_B=K_avoid*K_brake, where the three quantities each take Boolean values ​​1 and 0 and “1” means fulfilled.

[0051] In addition, a driving path criterion can advantageously be applied, which takes into account partial overlaps of the vehicle with object 8 or 18.

[0052] If the trigger criterion K_B is met in step St8, the brake warning signal S4 and / or brake control signal S2 is output in step St9.

[0053] Furthermore, it can be specified that in step St9, even if it is determined that a collision will occur even if emergency braking is initiated immediately, corresponding brake control signals S2 are output to the brake control devices 5 in order to reduce the impact force by the emergency braking.

[0054] To determine the avoidance trajectory 10, instead of the Fig. 1, a sliding sine function in polar coordinates can also be applied. This is preferably done when cornering, where the avoidance trajectory is transferred from one curve radius R1 to another R2, which corresponds to a change of lane. The curve radii R1 and R2 each have the same origin as a reference point, i.e., with a distance r from the origin and a polar angle Phi. The difference between the curve radii corresponds to the lateral offset: R1-R2 = dy and the arc length along the radius R1 represents the longitudinally available avoidance space L. The curve radius R1 is now guided over the angle phi to the curve radius R2 using the avoidance trajectory: ⇒r(phi)=dyLphi∗R1−dy2πsin(2πLphi∗R1)+R1,phi∈0…LR1 r(0)=R1 r(LR1)=dy+R1=R2

[0055] In the situation of Fig. 1 and Fig. 2, several detected objects 8, 18 can also be combined as a common object space 26 when determining the respective avoidance trajectory 10, 11. This is shown in Fig. 1 by objects 8 and 8a indicated by dashed lines, which are thus evaluated as a common object space 26, so that the avoidance trajectory 10 or 11 is placed around the common object, in Fig. 1 e.g. to the left and not to the right.

[0056] If vehicle 1 is equipped with an automatic steering or steering control device, an automatic evasive maneuver can also be determined.

[0057] Fig. Figure 6 shows a diagram in which the relative distance dx of the vehicle and the object 8 or 18 is plotted against the corresponding relative speed dv. Here, a curve L10 shows the minimum distance L1 in the case of Fig. 1, ie the avoidance trajectory 10 for the lane change and a curve L11 the minimum distance L2 in the case of Fig. 2, i.e., the avoidance trajectory 11 for cornering; furthermore, a curve Lbr is drawn for a minimum braking distance at which emergency braking of vehicle 1 just barely avoids a collision. Of particular interest here are the marked intersection points dx1, dv1 and dx2, dv2 of curves L10 and L11 with Lbr. Generally, L11 runs below L10; dx1 < dx2 and dv1 < dv2 apply.

[0058] Thus, for the avoidance trajectory 10 the Fig. 1 a collision can be avoided by braking until dv2, since the curve Lbr runs below L10 until then; accordingly, for the avoidance trajectory 11 the Fig. 2 to dv1 a collision can be avoided by braking, since the curve Lbr runs below L11 until then.

[0059] To initiate braking, both the braking criterion K_brake and the respective avoidance criterion K_avoid must be met. Since, according to the described embodiment, only braking is applied and automatic steering intervention is not provided, the avoidance criterion K_avoid must be met before the braking criterion K_brake to prevent an accident. Otherwise, it would already be too late for braking, and only the severity of the accident can be mitigated. With the avoidance trajectory 10 for the lane change, braking can still be applied to prevent an accident at higher relative speeds dv, since avoiding the accident is no longer possible earlier. List of reference symbols (part of the description) 1 vehicle 2 lanes 2a, 2b, 2c, 2d lanes 3 distance sensors 4 Control device 5 brake control devices 6 storage 7 Navigation device 8 moving object (another vehicle) 8a another moving object 10 Avoidance trajectory for lane changes 11 Avoidance trajectory for cornering 18 stationary object (traffic sign) 20 Display device 22 Emergency braking system 26 common object space F Direction of travel v Driving speed ax longitudinal acceleration q Lateral acceleration ω yaw rate t time t0 current time tz previous point in time tz to t0 past period x x-direction, longitudinal direction, Cartesian coordinate y y-direction, transverse direction, Cartesian coordinate v8 driving speed a8 Longitudinal acceleration of the object 8 q8 Lateral acceleration of the object 8 dvy8 lateral speed dx8 Distance in longitudinal direction dy8 cross offset S1 environmental sensor signals S2 brake control signals S3 Driving dynamics status signals S4 Brake warning signal x_w turning point x_q1 first extreme value x_q2 second extreme value dy lateral offset dx distance dvy lateral relative velocity dvx relative longitudinal speed dax relative longitudinal acceleration L1 Minimum distance for avoidance trajectory 10 in Fig. 1 L2 Minimum distance for avoidance trajectory 11 in Fig. 2 K_avoid avoidance criterion K_brake braking criterion K_B Trigger criterion for braking v_m measurement limit q_g lateral acceleration limit R1, R2 curve radii in avoidance trajectory 10 R3 Curve radius in avoidance trajectory 11 phi angle Lbr curve for minimum braking distance L10, L11 Curves for avoidance trajectories 10, 11

Claims

[1] Method for determining a trigger criterion (k_B) for outputting brake signals (S4, S5) in a vehicle (1), comprising at least the following steps: Detecting at least one object (8, 18) in the vicinity of the vehicle (1) (St1), Determine whether the vehicle (1) is on a collision course with the object (8, 18) (St2), when determining a collision course with the detected object (8, 18), checking an avoidance criterion (K_avoid) which is met if no avoidance maneuver for the vehicle (1) can be determined or is possible (St3), at least if the avoidance criterion is met, check a braking criterion (K_brake) (St8), the object (8, 18) is classified into at least a first or a second class depending on its properties (St3), where the evasive maneuver of the vehicle (1) is determined as a function of the classification of the object (8, 18) in such a way that, when classified in a first class, a first evasive trajectory (10) is determined and, when classified in a second class, a second evasive trajectory (11) different from the first evasive trajectory is determined (St4, St5, St6, St7), where the avoidance criterion (K_B) for issuing brake signals (S2, S4) is met if the avoidance criterion (K_avoid) and the brake criterion (K_brake) are met, characterized by that at least one extreme value (x_q1, x_q2, q(v,r)) of the lateral acceleration (q) acting on the vehicle is determined from the determined first or second avoidance trajectory (10, 11) and as an avoidance criterion (K_avoid) the at least one extreme value (x_q1, x_q2, q(v,r)) is compared with a lateral acceleration limit value (q_g), wherein the avoidance criterion (K_avoid) is met if the at least one extreme value (x_q1, x_q2, q(v,r)) is greater in magnitude than the lateral acceleration limit value (q_g). [2] Method according to claim 1, characterized by that the braking criterion (K_brake) indicates whether an emergency braking operation is necessary to prevent a collision with the detected object (8, 18), and if the avoidance criterion (K_avoid) is met and the braking criterion is met, the braking signal (S4) is output as a braking control signal (S2) or braking warning signal (S2) to prevent the detected collision or to reduce the severity of the collision. [3] Method according to claim 1 or 2, characterized bythat a driving path criterion is additionally determined and the trigger criterion (k_B) is only fulfilled if the driving path criterion is also fulfilled. [4] Method according to one of the preceding claims, characterized by that in the classification of the detected object (8, 18) its detected geometric properties (dy) and / or dynamic properties (d8) in a previous detection period (tz- t0) are used, wherein at least the driving speed (v8) of the object (8, 18) or the acceleration of the object is used as the dynamic driving property and wherein at least one of the following parameters is used as the geometric properties: shape of the object (8, 18), size or extent (dy) of the object (8, 18), position of the object (8, 18). [5] Method according to one of the preceding claims, characterized by that the object (8,18) is classified into the following classes: - stationary object (18) which has not shown any driving speed during the detection period, - moving or stationary object (8) to which a driving speed (v8) above the measuring limit value (v_m) was assigned at least once during the detection period (tz-t0). [6] Method according to claim 5, characterized by that a moving or stationary object (8) is divided into - a currently moving object (8) with current speed (v8) above the measurement limit value (v_m), and - a stationary object (8) which is currently stationary and at least temporarily has a driving speed (v8) above the measurement limit value (v_m) during the detection period, wherein identical or similar avoidance trajectories (10) are formed for the classifications of the object (8) as stationary or moving. [7] Method according to claim 5 or 6, characterized bythat when the detected object is classified as a moving or stationary object (8), the avoidance trajectory (10) is determined as a lane change or lane change trajectory of the vehicle (1). [8] Method according to one of claims 5 to 7, characterized by that when the detected object is classified as a moving or stationary object (8), an avoidance trajectory (10) is determined in which a final orientation of the vehicle corresponds within a tolerance range to the current orientation of the vehicle (1). [9] Method according to one of claims 6 to 8, characterized by that when the detected object is classified as a stationary object (18), an avoidance trajectory (11) is determined with a curve radius (R3) which is constant at least in one end region within a tolerance range. [10] Method according to one of the preceding claims, characterized bythat on the basis of map data / navigation data it is determined whether a lane course of the avoidance trajectory (10) of the vehicle (1) is possible based on the map data. [11] Method according to one of the preceding claims, characterized by that a relative avoidance trajectory (10, 11) of the vehicle (1) is determined, which represents a change in the avoidance trajectory compared to a projected own trajectory, wherein the projected own trajectory is determined as an extrapolation of an own trajectory traversed in a previous evaluation period (tz - t0). [12] Method according to claim 11, characterized by that at least one of the following quantities is determined in polar coordinates with an angle variable and a radius variable: relative avoidance trajectory, avoidance trajectory, projected own trajectory, own trajectory, driving dynamics properties of the vehicle, driving dynamics properties of the detected object, driving path, lane change, driving path change. [13] Method according to one of the preceding claims, characterized by that the brake signals are brake display signals (S4) for warning a driver of the vehicle and / or brake control signals (S2) for controlling brake actuating devices (5) to initiate automatic braking. [14] Method according to one of the preceding claims characterized by that detected objects (8, 18) are classified before checking whether they are on a collision course. [15] Method according to one of the preceding claims, characterized bythat several adjacent objects (8, 18) are combined to form a common object space (26), wherein a distance between the adjacent objects is less than a width or scaled width of the vehicle (1). [16] Method according to one of the preceding claims, characterized by that a minimum distance (L1, L2) at which the avoidance criterion (K_avoid) is met is smaller when the detected object is classified as a stationary object (18) than when the detected object is classified as a moving object (8) or a standing object. [17] Emergency braking system (22) for a vehicle (1), in particular for carrying out a method according to one of the preceding claims, comprising: an environmental sensor system (3) for detecting at least one object (8, 18) in the surroundings of the vehicle (1), a control device (4) which is provided - to determine whether the vehicle (1) is on a collision course with the detected object (8, 18), - when determining a collision course with the detected object (8, 18) to check an avoidance criterion (K_avoid) which is met if no avoidance maneuver for the vehicle (1) can be determined or is not possible, - at least when the avoidance criterion is fulfilled to check a braking criterion (K_brake) and output brake signals (S4, S5) depending on the check of the braking criterion, - wherein the control device (4) classifies the object (8, 18) into at least a first or a second class depending on its properties, wherein the evasive maneuver of the vehicle is determined as a function of the classification of the object (8, 18) in such a way that, when classified in a first class, a first evasive trajectory (10) is determined and, when classified in a second class, a second evasive trajectory (11) is determined which differs from the first evasive trajectory, characterized by in that the control device (4) is designed to determine at least one extreme value (x_q1, x_q2, q(v,r)) of the lateral acceleration (q) acting on the vehicle from the determined first or second avoidance trajectory (10, 11) and to compare the at least one extreme value (x_q1, x_q2, q(v,r)) with a lateral acceleration limit value (q_g) as an avoidance criterion (K_avoid), the avoidance criterion (K_avoid) being met if the at least one extreme value (x_q1, x_q2, q(v,r)) is greater in magnitude than the lateral acceleration limit value (q_g). [18] Vehicle, in particular commercial vehicle, with an emergency braking system (22) according to claim 17.

Images