WO2015188904A1 - Method and device for at least reducing the consequences of a collision of a vehicle - Google Patents

Abstract

The invention relates to a method for at least reducing the consequences of a collision of a vehicle (F). According to the invention, at least one collision avoidance operating mode (K1 to K3) from a set of at least three different collision avoidance operating modes (K1 to K3) is activated. A braking maneuver is executed on activation of a first collision avoidance operating mode (K1), an evasion maneuver is executed on activation of a second collision avoidance operating mode (K2), and a braking maneuver and an evasion maneuver are simultaneously executed on activation of a third collision avoidance operating mode (K3). The respective collision avoidance operating mode (K1 to K3) is activated depending on a vehicle speed (v_ego) of the vehicle (F), a traffic situation, and/or an available free space in a vehicle environment. The invention further relates to a device (V) for at least reducing the consequences of a collision of a vehicle (F).

Description

 Method and device at least for reducing the consequences of a collision of a

 vehicle

The invention relates to a method at least for reducing the consequences of a collision of a vehicle. Furthermore, the invention relates to a device at least for

Mitigate the consequences of a collision of a vehicle.

DE 10 2012 001 405 A1 discloses a method for carrying out a

Lane change of a vehicle from a Momentanfahrspur to an adjacent target lane described. This is a transition trajectory of the

Momentanfahrspur in the destination lane, taking into account given

Limits determined for vehicle dynamics parameters.

Furthermore, DE 10 2005 003 274 A1 proposes a method and a device for avoiding and / or reducing the consequences of collisions when avoiding a vehicle in front of obstacles, wherein by means of at least one

Environment sensor at least one obstacle is detected and data of the obstacle are determined. Further, on the basis of the data of the obstacle as well as the data of the vehicle, one favorable in support of an avoidance operation

Vehicle deceleration determined and delayed the vehicle accordingly.

In addition, DE 10 2011 106 520 A1 discloses a method for preventing a collision of a vehicle. This is an ongoing operation of a

Detects brake pedals of the vehicle and it is closed to a collision hazard when the brake pedal is pressed quickly. Furthermore, a position and a movement of objects in a vehicle environment relative to the vehicle are detected and assessed with regard to the risk of collision. If a collision hazard is detected, it will automatically turn on

Evasive maneuvers of the vehicle carried out when evaluating the

Collision hazard results in that a collision can be prevented by an evasive maneuver, but not by a braking maneuver. The invention is based on the object to provide a comparison with the prior art improved method and an improved device at least to reduce the consequences of a collision of a vehicle.

With regard to the method, the object is achieved according to the invention with the features specified in claim 1 and with respect to the device with the features specified in claim 10.

Advantageous embodiments of the invention are the subject of the dependent claims.

In a method for at least reducing the consequences of a collision of a vehicle, it is provided according to the invention that at least three of them are used

various collision avoidance modes at least one

Collision avoidance mode is activated, wherein upon activation of a first collision avoidance mode, a braking maneuver is performed, when activating a second collision avoidance mode, an evasive maneuver is performed, and wherein upon activation of a third collision avoidance mode simultaneously a braking maneuver and an evasive maneuver is performed. The respective collision avoidance operating mode is activated as a function of an own driving speed of the vehicle, a traffic situation and an available free space in a vehicle environment.

The method enables a collision avoidance or at least a reduction of consequences of an impending collision for a partially or highly automated driving. Thus, if an impending collision can no longer be prevented by a sole braking maneuver, the collision may possibly be prevented by the avoidance maneuver, provided that a collision-free clearance is available in a neighboring lane.

The first collision avoidance mode, i. H. the initiation of a braking maneuver is activated when the own driving speed of the vehicle is below a predefinable limit speed, so that an imminent collision by means of a braking maneuver is considered to be avoidable.

For the first collision avoidance mode is a Bremstrajektorie with a minimum allowed negative longitudinal acceleration or with a maximum approved longitudinal deceleration determined by means of which by the braking maneuver a collision is still avoidable.

 If the own driving speed of the vehicle exceeds the limit speed, the second becomes dependent on the traffic situation and the free space

Collision avoidance mode or the third collision avoidance mode is activated. Thus, an imminent collision may possibly be prevented by an alternative or additional evasive maneuver.

The third collision avoidance mode, i. H. the initiation of a braking and avoidance maneuver is activated when, by means of the first collision avoidance mode, a collision is unavoidable and an available free space exists. This can be prevented by means of braking and avoiding an imminent collision without other vehicles may be damaged.

After completion of the avoidance maneuver, the first collision avoidance mode is activated, so that the vehicle comes to a stop in the available free space and thus remains there.

In the method, the vehicle environment is divided into surrounding areas, wherein an object recognition and a detection of available free spaces is carried out within each surrounding area. The detection of the surrounding areas by means of an environment detection unit, which, for example, on and / or arranged in the vehicle sensors, for. As a camera based.

Furthermore, the method carries out a collision check for each free space, by means of which free space status information is determined. The free-space status information indicates whether or not the particular free space considered is collision-free due to a constantly performed movement prediction.

In this case, the collision check is carried out taking into account a movement prediction, wherein in the movement prediction movements of further vehicles and obstacle objects on a current lane of the own vehicle as well as on at least one adjacent lane are preconditioned. This allows a precise, Determined on the current traffic situation determination of the corresponding collision avoidance mode.

The invention further provides a device at least for reducing the consequences of a collision of a vehicle, wherein according to the invention from a set of at least three different collision avoidance modes at least one collision avoidance mode can be activated, wherein upon activation of a first collision avoidance mode, a braking maneuver is feasible , in the activation of a second collision avoidance mode, an evasive maneuver is feasible and at the same time a braking maneuver and an evasive maneuver is feasible when activating a third collision avoidance mode. The respective

Collision avoidance mode is dependent on its own

Driving speed of the vehicle, a traffic situation and an available free space in a vehicle environment activatable.

The device improves the safety of the vehicle over the prior art, since a collision avoidable and / or at least following a collision can be reduced by means of the device.

Embodiments of the invention are explained in more detail below with reference to drawings.

Showing:

Fig. 1 shows schematically a device at least for the reduction of consequences of a

 Collision of a vehicle,

2 shows a flow chart for carrying out a method for reducing

 Consequences of a collision of a vehicle,

3 schematically shows a first exemplary driving situation with four vehicles and three lanes of a roadway,

4 shows a path-time diagram for predicted vehicle movements of the first exemplary driving situation shown in FIG. 3, 5 schematically shows a second exemplary driving situation with four vehicles and two lanes of a roadway,

6 shows a path-time diagram for predicted vehicle movements of the second exemplary driving situation illustrated in FIG. 5,

7 schematically shows a third exemplary driving situation with four vehicles and two lanes of a roadway,

8 shows a path-time diagram for predicted vehicle movements of the third exemplary driving situation illustrated in FIG. 7, FIG.

FIG. 9 schematically shows a fourth exemplary driving situation with further vehicles and three lanes of a lane, whereby no collision-free clearance is available, FIG.

Fig. 10 shows schematically a fifth exemplary driving situation with other vehicles and three lanes of a road and

11 schematically shows a flowchart for a search strategy for determining a comfortable lateral acceleration.

Corresponding parts are provided in all figures with the same reference numerals.

FIG. 1 shows a device V, at least for the reduction of consequences of a collision of a vehicle F shown in FIGS. 3, 5, 7, 9 and 10, in a schematic representation.

The device V comprises a plurality of components, which are described in more detail below.

For detecting a position of a current lane and a position of the vehicle F, a bordautonomome localization device 1 is provided. The Localization device 1 is, for example, a video-based localization system, which searches for roadway markings on the roadway and / or landmarks by means of cameras mounted in vehicle F. An output signal of

Localization device 1 is a first output vector Y_L.

The device V further comprises a localization and navigation device 2 with detailed map information for determining an intrinsic position of the vehicle F and for determining location positions and a type of traffic sign,

Traffic lights, stop lines, roundabouts, which are in a predefinable

Advance distance on a planned route before the current own position of the vehicle F are located. This is z. B. by means of a global

 Positioning system, short: GPS, and a high-precision electronic map with detailed information about traffic signs, traffic lights, stop lines, etc. possible. An output signal of the localization and navigation device 2 is a second output vector Y_N.

Furthermore, the device V comprises a road vehicle and vehicle vehicle communication device 3 for determining a road condition, for. B. on the road, in a pre-distance on the planned route, to determine the location positions of traffic obstructions, eg. B. stagnant vehicles or a congestion end location position, which are in a pre-distance on the planned route in front of the current own position of the vehicle F. An output signal of the roadway vehicle and vehicle vehicle

Communication device 3 is a third output vector Y_K.

In addition, the device 1 includes a traffic sign and

Lichtsignalanlagenzustandserkennungseinheit 4, which on a

Stereo video camera system with downstream image analysis based and which at least the location positions and the type of traffic signs, the location positions of holding and waiting lines and the location positions and the current state of

Traffic light systems in a Vorausentfernung on the planned route in front of the own vehicle F determined. The output signal of traffic signs and

Light signal state detection unit 4 is a fourth output vector Y_S. Furthermore, an environment detection unit 5 is arranged, which is based on a detection of a reflection of emitted electromagnetic radiation, for. As radar, lidar, for detecting a drivable space around the own vehicle F based. The

Output signal of the environment detection unit 5 is a fifth output vector Y_U.

In addition, the device V comprises an object recognition unit 6 which is connected to radar systems and / or stereo video camera systems downstream

Image evaluation based, which determines at least one longitudinal and one lateral position of the obstacle objects and persons surrounding the vehicle F relative to the own vehicle F and / or lane. The output signal of

Object recognition unit 6 is a sixth output vector Y_0.

Furthermore, a driving condition detection unit 7 is arranged, which is an actual driving condition consisting of driving dynamics variables, such. B. own

Driving speed v_ego, an acceleration a, a lateral acceleration a_y, a yaw rate psip, a steering angle delta, and a coefficient of friction mue determined. The output signal of the driving state recognition unit 7 is a seventh output vector Y_F-

The output signals of the above-described components of the device V are transmitted to a situation evaluation and decision unit 8 which includes a longitudinal collision evaluation unit 8.1, a lateral collision evaluation unit 8.2, a traffic situation determination unit 8.3 and a collision avoidance mode determination unit 8.4

The situation evaluation and decision unit 8 evaluates, depending on the output signals transmitted thereto, based on the information about the position and location of the own vehicle F on the road, the location positions or distances, and the type of road signs, the location positions, or distances, of hold and waiting lines as well as the location positions or distances, and the current states of

Traffic light systems, the environment information about the free traffic space around the own vehicle F around and the information of the driving condition as well as the information about the traffic obstacles surrounding the own vehicle F and the information of the driving condition. Furthermore, the situation evaluation and decision unit 8 for a longitudinal dynamics trajectory planning unit 9 determines the input variables required for braking or acceleration of the own vehicle F, ie the limit values characterizing the driving dynamic behavior, eg. B. during a driving maneuver when driving a maximum permitted positive longitudinal acceleration ax_max and braking the vehicle F a negligible negative longitudinal acceleration ax_min or a maximum permissible longitudinal deceleration and depending on the traffic obstacle the target driving speed v_z _ie i, which is equal to zero for stationary traffic obstructions and with moving traffic obstruction objects equal to one

Obstacle object speed vobj.

In addition, the situation evaluation and decision unit 8 optionally determines the input variables necessary for avoiding a transverse dynamics trajectory planning unit 10, ie. H. the driving dynamic behavior characteristic limit values, z. B. a maximum allowed lateral acceleration ay_max or a maximum

permitted yaw angular velocity and an evasion target lateral offset yAW_ ^ _ie |.

The longitudinal dynamics trajectory planning unit 9 generates state variables, such. B. a trajectory longitudinal acceleration a_trj, a trajectory driving speed v_trj and optionally a Trajektorien-Wegposition or a trajectory longitudinal distance s_trj as set values for a longitudinal control unit 11th

The transverse dynamics trajectory planning unit 10 generates state variables, such. A trajectory curvature c_trj, a trajectory heading angle psi_trj and a

Trajectory lateral position y_trj to achieve the evasion target lateral offset

as nominal values for a transverse control unit 12.

The longitudinal control unit 11 regulates the longitudinal dynamics trajectory prescribed by the longitudinal dynamics trajectory planning unit 9 and outputs as output signals a drive control signal u_m and a brake control signal u_br to a driver safety device 13.

The transverse control unit 12 regulates the transverse dynamics trajectory predetermined by the transverse dynamics trajectory planning unit 10 and outputs as an output signal a steering control signal ujenk to the driving safety device 13. The driving safety device 13 serves for driving stabilization and is, for example, an anti-lock braking system, in short ABS, an electronic stabilization program, ESP for short, anti-slip control, in short ASR, and / or a pre-safe brake. The

Driving safety device 13 performs three actuators 14 to 16 of the device V depending on the driving situation driving dynamics secured actuation signals and transmitted in a full braking with ABS a minimum possible negative

 Longitudinal acceleration ax_min_ABS, a maximum transverse acceleration ay_max_ABS possible during full braking with ABS, and a maximum longitudinal acceleration ax_max_ASR possible with full drive with ASR to the situation evaluation and decision unit 8.

The control signals of the driving safety device 13 are another drive control signal u_m_ABS, another brake control signal u_br_ABS and another

Steering position signal u_lenk_ESP. Return variables to the situation evaluation and decision unit 8 are the minimum permitted negative longitudinal acceleration during full braking with ABS ax_min_ABS, the maximum possible lateral acceleration during full braking with ABS ay_max_ABS and the maximum possible longitudinal acceleration at full acceleration with ASR ax_max_ASR.

In addition, the device V comprises three actuators 14 to 16, in particular a first actuator 14, which serves to drive a drive of the vehicle F, a second actuator 15, which serves to control a brake of the vehicle F, and a third actuator 16, which Control of a steering of the vehicle F

is provided.

In addition, the device V comprises an operating and display unit 17, which informs the driver of an imminent collision and which for the driver

Operating instruments for switching on and off a partially or highly automated

Driving function provides. The output signal of the operating and display unit 17 is an operating signal x_b, which is transmitted to the situation evaluation and decision unit 8. The input signal of the operating and display unit 17 is a

Display signal x_a, which is transmitted from the situation evaluation and decision unit 8 to the control and display unit 17. As previously described, the situation assessment includes and

Decision unit 8, the longitudinal collision assessment unit 8.1, the

Cross collision assessment unit 8.2, the traffic situation determination unit 8.3 and the collision avoidance mode determination unit 8.4.

The longitudinal collision evaluation unit 8.1 evaluates and checks for freedom from collision in the vehicle longitudinal direction by means of prediction of the vehicle movements, in particular for obstacle objects, for B. slower moving, retarding or standing

Advance vehicles, in the current lane, z. B. in driving situations with a jam end.

The lateral collision assessment unit 8.2 monitors, evaluates and checks the environment around the own vehicle F for collision freedom, in particular for

Obstacle objects or slower moving, decelerating or standing vehicles in front and faster moving following vehicles on a left and right neighbor lane and selects in case of collision freedom suitable for collision avoidance target lane, as will be described later.

The traffic situation determination unit 8.3 evaluates those of the location and navigation device 2, of the environment detection unit 5 and of the

Object detection unit 6 as well as the roadway vehicle and vehicle-vehicle communication device 3 provided information and assigns the current driving situation of a predetermined class or a type of

Traffic situations, z. For example, a jam end on a highway, a following trip on a highway, a following trip on a country road, an intersection on a country road, a subsequent trip in a village and / or a crossroads in a village, etc.

In the collision avoidance mode determination unit 8.4, depending on the own vehicle speed and depending on the results of the

Collision freedom checks of the longitudinal collision assessment unit 8.1 and the collision collision assessment unit 8.2 as well as dependent on the

Traffic situation determination unit 8.3 certain traffic situation type the

Collision avoidance mode determines, as will be described later. FIG. 2 shows a flowchart for carrying out a method for reducing consequences of a collision of a vehicle F realized in the collision avoidance mode determining unit 8.4.

The flowchart is a state machine in which three

Collision avoidance modes for collision avoidance first

Collision avoidance mode K1, as a second collision avoidance mode K2 and as a third collision avoidance mode K3 and with an additional one

In the first collision avoidance mode K1, a braking maneuver is initiated, in the second collision avoidance mode K2 an evasive maneuver is initiated and in the third collision avoidance mode K2

Collision avoidance mode K3 is simultaneously initiated a braking and evasive maneuver.

After the start or initialization, the first collision avoidance mode K1 is always assumed. The transition from one of the collision avoidance modes K1 to K3 to another of the collision avoidance modes K1 to K3 is dependent on a number of transition conditions A, B, C, D. The

Transition conditions A, B, C, D are dependent on the current own

Driving speed v_ego, the type of current traffic situation, from

Presence of free spaces and whether existing free spaces are collision-free due to the movement prediction performed in the longitudinal collision evaluation unit 8.1 and in the collision collision assessment unit 8.2. In particular, only the first one is below a limit traveling speed for collision avoidance

Collision avoidance mode K1 set.

Above this limit driving speed, depending on the driving situation, only the second or third collision avoidance operating mode K2, K3 is changed if a collision-free free space for avoiding exists. For a first

Transition condition A and a third transition condition C must therefore be met in principle the condition that own driving speed greater than the

Limit speed is. Therefore: v_ego> v_ego_grenz where v_ego describes the own driving speed and v_ego_grenz the limit speed.

For a second transition condition B and a fourth transition condition D, d. H. for transitions back to the first collision avoidance mode K1, an already activated avoidance operation must be in a target lane or in a target corridor

be completed. This is the case when, after an evasive action, the own vehicle F is completely in the destination lane or in a destination corridor. Then the original destination lane becomes the current current lane.

For the first transition condition A shown in FIG. 2, for example, the following condition can be defined for the transition to the third collision avoidance mode K3:

(v_ego> v_ego_limit)

AND (Status_FR_V = Not collision free with EGO full braking (NKF))

 AND ((FR_ziel = FR_V_re) OR (FR_ziel = FR_V_re))

 AND (Status_FR_ziel = collision free KF)

 AND (type traffic situation = congestion on freeway), where FR_V is a front clearance, i. H. a free space located in front of the own vehicle, and FR_ziel describes a target clearance to be traveled to avoid a collision.

For the third transition condition C can for the transition to the second

Collision avoidance mode K2 be set, for example, the following condition:

(v_ego> v_ego_limit)

 AND (Status_FR_V = collision free with EGO full braking (KF))

AND ((FR_ziel = FR_V_re) OR (FR_ziel = FR_V_re))

AND Status_FR_ziel = collision free KF)

AND (type_trip situation = follow-on on motorway), where type_transport situation describes a type of a current traffic situation.

The first and third transition conditions A, C can thus be new for everyone

additional type of traffic situation or for elements of the class of

Traffic situations are easily extended.

Figure 3 shows schematically an exemplary driving situation with the vehicle F and three other vehicles F1 to F3 and three lanes.

Here, the own vehicle F is on a middle lane, which is a current lane of the own vehicle F. In front of the own vehicle F there are on the middle lane a first further vehicle F1, on a left lane a second further vehicle F2 and on a right lane a third further vehicle F3.

Depending on the driving speeds of the three other vehicles F1 to F3 and their own driving speed v_ego, the driving situation shown is dependent on either the traffic situation type "traffic jam on highway" or the traffic situation type "follow-on drive on motorway".

For the assessment of the situation and the decision, those of the

 Environment detection unit 5 detected areas in the present embodiment as surrounding areas according to in front environment areas UB_V, UB_VJi, UB_V_re and in the back environment areas UB_H, UB_H_li, UB_H_re divided, which the

For simplicity, are shown rectangular.

For each of these environment areas UB_H, UB_V, on the basis of the information of the environment detection unit 5, an environment status information, hereinafter also abbreviated as StatusJJB, is determined whose content indicates whether objects are within this range. Thus there is as information either StatusJJB = object free, short: OF, or StatusJJB = not object free, short: NOF. Within each environment area UB_H, UB_V is constantly checked whether there are objects and if there are free spaces. The free spaces are correspondingly divided into front free spaces FR_V, FR_V_li, FR_V_re and into back free spaces FR_H, FR_H_li, FR_H_re. If there is no object in an environment, the free space in the longitudinal direction is the same as the surrounding area. The size of a clearance in the transverse direction should at least include the size of the width of the own vehicle B_ego. If lanes with lane widths are present in the surrounding area that are greater than or equal to the width of the driver's own vehicle F, the free space in the transverse direction can be limited by the respective neighboring lane width.

For each of the front clearances FR_V, FR_V_li, FR_V_re and the rear clearances FR_H, FR_H_li, FR_H_re, taking into account the in the

Longitudinal collision evaluation unit 8.1 and in the lateral collision assessment unit 8.2 constantly performed motion prediction made a collision check while a free space status information, short: Status_FR, determines its content indicates whether the respective considered free space due to the constant movement prediction is collision-free or not. Thus, there is free space status information for each free space, which provides either the status "status_FR = collision free, short: KF" or the status "status_FR = non collision free, short: NKF".

In the driving situation shown in FIG. 3, the rear clearances FR_H, FR_H_li, FR_H_re are the same length as the rear surrounding areas UB_H, UB_H_li, UB_H_re. In this driving situation illustrated by way of example three front clearances FR_V, FR_V_li, FR_V_re are present, wherein the own vehicle F in the current lane the front clearance FR_V, located on the left lane second further vehicle F2 the left front clearance FR_V_li and on the right third lane third vehicle F3 limit the right front clearance FR_V_re. For each of the limited free spaces, in this case the front clearances FR_V, FR_V_li, FR_V_re, the collision freedom is checked in the longitudinal collision assessment unit 8.1.

For the driving situation exemplified in FIG

Traffic situation determination unit 8.3 depending on the driving speeds of the other three vehicles F1 to F3, the own driving speed v_ego and possibly information of the road vehicle and vehicle-vehicle communication device 3 the traffic situation type either to "traffic jams on the highway" or to "follow on highway". The longitudinal collision evaluation unit 8.1 assesses and checks for collision freedom in the vehicle longitudinal direction in principle by means of prediction of the vehicle movements. In this case, the movements of the other vehicles F1 to F3 and obstacle objects in the current lane and in at least two neighboring lanes are predicted, for example slower moving, decelerating or standing vehicles ahead and faster driving

Follow-up vehicles especially in driving situations with jam end.

For the longitudinal dynamics prediction of the movements or relative movements of a preceding vehicle (i) and a following vehicle (j), the designations for the distances traveled s_,, Sj and for the distance from the vehicle in front (i) to the following vehicle (j) d_ _u introduced. Sj - _u with the initial distance D0_ D0_ = _u + s_ _f thus results for the distance d_ _u.

The collision assessment depends on a specific time horizon

predicted driving behavior of the following vehicle (j) and of the vehicle in front (i).

For an unrestrained, with constant driving speed Vj driving following vehicle 0) are known to be in the vehicle following task, the following cases:

In the event of an unrestrained slowly with constant speed (v_ _f <vj) propelled forward vehicle (i) results in a Berühr-time distance or collision interval, also time to collision, in short: TC called to

and the sequence time interval usually used for a following trip

In the event that the unbraked vehicle ahead (i) stationary at the same constant speed as the following vehicle (j), ie ( _j = _j ) drives, threatens no

_{Danger of} collision and the collision time interval T _{CJJ approaches} infinity. The sequence time interval usually used as the basis for a follow-up run is then given ^" you ^"" you ^"

 _V

In the case of a stationary vehicle in front of the vehicle (i), the contact time interval or collision time interval and also the sequence time interval usually assumed for a subsequent journey result

The following will be different for a braked follower vehicle ()

Behavior cases of the vehicle in front (i) considered, as in particular

Congestion-driving situations can occur.

For the real at a time t = t _k conducted longitudinal dynamics prediction of a leading vehicle (i) and a following vehicle (j) as the initial values of the initial distance D0_ _u = D0_ _u (t _k) and the initial speeds vO _j, voj based placed. In this case, one's own vehicle F may be the following vehicle depending on the object situation) or also the preceding vehicle (i).

 In the following motion prediction for determining the relevant parameters for the longitudinal collision, z. As collision time interval, standstill braking distances, the model equations are presented as an example in the event that the own vehicle F is the follower vehicle (j).

Here, the own initial driving speed v0_ _eg o = vO_eg ₀ (t _k ), the own negative longitudinal acceleration is to be braked with the is a_ _eg o = a_ _ego (tk), the time delay T _R with which the own vehicle F brakes, the initial running speed vo _j = v0 _j (t _k) and the negative longitudinal acceleration at which the vehicle looked advance (i) brakes a_ used.

For any real time t = t _k , the movements of the

Vehicles F, F1 to F3 with a prediction time τ for a prediction horizon [0, r _max ] predicted. In general, for any real point in time t = t _k with a braking trajectory f _{br g0} for the own vehicle F and an assumed brake _trajectory f _brJ for the considered _preceding vehicle (i), the predicted distances or braking distances to s_e _ga ( ^tk _> Ό = s _br _ _ega (tk, r) = f _hr _ _eg0 (r, v (. a_, TR) and s_, (tk, r) = s _{hr i} (h, τ) = f _hr _, (r, v0_ , ■).

The following exemplary model equations form the basis for the motion prediction performed in the longitudinal collision evaluation unit 8.1. In this case, it is assumed that the vehicle in question (i) either moves at a constant driving speed or brakes with a constant negative acceleration or is already stationary at detection. Furthermore, as a braking trajectory for the own vehicle F, a constant negative longitudinal acceleration a_ego = is assumed to be constant.

The predicted relative braking distance of the braked vehicle F to a moving, z. B. braking or slow moving, vehicle ahead (i) results for the prediction _interval [0, T _ssj ], ie as long as the vehicle _ahead (i) in the prediction has not come to a standstill yet

(1a)

^S br_ go_l

T + j [a »-fl ^lj ] - f + | α_" · 7 * ² } and for the prediction _interval [T _ssj , T _ss _ _eg o], ie after the preceding vehicle (i) has already come to a standstill in the prediction is to

(1b)

^s f, _r _e _g _, (t ^k , T) = s _{hr ega} (tk, r) - 5 _Är _, (tk, r) + - «<r ² + γ. ^■ TR

The predicted relative braking distance path of the braked vehicle F, in this case the follower vehicle to a stationary already Prädiktionsbeginn preceding vehicle (i) is obtained for the prediction interval [0, T _ss _ _e g _0], that is not as long as the own vehicle F in the prediction still has come to a standstill, too (1c) s _{hr ego i} (h, T)

τ ² +} ö ,, « ² j.

In this case, the relative braking distance also corresponds to the absolute

Braking distance or the resulting from the Bremstrajektorie of the own vehicle F absolute braking distance s _{hr eg0} = f _hr _ _eg0 .

The predicted dynamic actual distance is obtained from the host vehicle F beginning with the distance dO _{J e} go = d0 _j _ _e g _O (t _k) from the preceding vehicle (i)

(2) DJ_ _EC o (t _k, T) = d0j_ _ego (t _k) - S _br _ego (tk, Γ) + S _br j (tk, 7) = d0j_ _e go (tk) "S _br _egoj (tk , Γ) ·

The predicted standstill time interval is for the delayed with time delay T _R own vehicle F:

For a braked front-end vehicle (i), the predicted standstill interval is:

The predicted relative braking distance of the braked own vehicle F to a braked vehicle ahead (i) at standstill of the vehicle F, also called relative stoppage braking distance, is:

( ^5a ) Sbr "ego i ( ^tk ) = ^S rc _g o ₁ 0 *, * =

The predicted relative braking distance of the braked vehicle F to a preceding vehicle (i) which is already stationary at the start of prediction when the vehicle F is stationary is: vO. "

(5b) s _hr _ _m _ _ego (tk) = s _{hr eg} " _j (t _k , τ = Ts, _ = v0.e ₈ " · TR +

In this case, the relative standstill braking distance also corresponds to the absolute standstill braking distance.

 The predicted standstill distance from the vehicle in front (i) to the vehicle F is

(6) d _S s_i_ego (tk) ^_ DJ_ _e g ₀ (t k, Γ - T _ss _ _e go) ⁼ DOJ ego (tk) 'St', r_ss_ego_i '

It should be noted that the standstill distance dss_i_ego is not always the smallest distance that occurs during a braking maneuver. From the derivative of the predicted dynamic actual distance d (r) = d _j _ _e go (t _k> r) is determined, the distance minimum. The distance minimum or the critical distance to the moving

Vehicle in front (i) is reached at the critical time (t _kr ) or at the critical time interval T _kr .

The predicted dynamic-critical time interval of the braked vehicle F to a moving vehicle ahead (i) results in

The predicted dynamic-critical time interval of the braked vehicle F with respect to a preceding vehicle (i) which is already stationary at the beginning of the prediction results in too

In this case, the dynamic-critical time interval corresponds to the standstill time interval for the vehicle F, d. H. the follower vehicle.

The predicted relative braking distance of the braked vehicle F to a moving vehicle ahead (i) at the dynamic critical time, also called dynamic critical braking distance, in which the moving vehicle ahead (i) and the moved vehicle F in the prediction have not come to a standstill, it follows:

The predicted relative braking distance of the vehicle F to one already

Prediction start stationary vehicle (i) at the dynamic critical time, d. H. the dynamic-critical braking distance, results in:

( ^8b ) ^S br eggo, ( ^tk ) = ^, ego, (tk, T = Tkr) = S _hr ego ₍ · (tk, T = Tss _ ego) = S _hr "s_ego., (Tk).

In this case, the dynamic critical braking distance is equal to the relative

Sti I Ista nd s-B remsweg route.

The predicted dynamic-critical distance results with the initial distance dO _j from the vehicle in front (i) to the vehicle F

(9) ^u kr_i_ego (t _k ) = dj_ _ego (t _k , r = T _kr ) = dOj_ _e go (t _k ) - s _br _ kr_ego. j (tk) ·

The predicted contact time interval or collision time interval of the braked

Vehicle F to a braked vehicle ahead (i) or to one already

Prediction start stationary vehicle ahead (i) results in the event that the initial distance is smaller than the relative standstill braking distance, d. H. For

dO_ _Le go (t _k ) <s _br _ ss_ego _i (tk), ZU

 vO,

 (10) with dO = dOj in a braked ahead vehicle (i)

 - Oj

and dO = dt at a vehicle stationary at the beginning of the prediction (i) The predicted contact time interval or collision time interval of the braked

Vehicle F to a moving, in particular braking or slower moving, preceding vehicle (i) results in the event that the initial distance is smaller than the dynamic-critical braking distance, ie for dO_ _L ego (tk) <Sb _r _kr_ego_i (k ), to:

 with a_, c = a_, - with a braked ahead vehicle (i)

 with a = - with an unrestrained, slower advance - vehicle (i)

The maximum value of the predicted relative braking distance results from the maximum of standstill braking distance and dynamic-critical braking distance to:

(12) s _ir _ _nl _ _{tM! L} (tk) = MAX (s _hr _ _nl _ ", s _br _ _nl _ _lr ).

Similarly, the corresponding predicted relative results

Braking distances and the characteristics relevant to the longitudinal collision in the event that the vehicle considered for the collision is a braked following vehicle (j) coming from behind and the own vehicle F is a braked vehicle ahead (i).

In principle, those relevant for longitudinal collision assessment can be used

Driving situation-dependent parameters from the equations (1) to (12) are also determined by prediction by means of numerical pre-simulation. This may be advantageous in more complicated braking trajectories because an analytical solution for determining the relevant characteristics is more difficult than a numerical one

Can be pre-simulation.

The method provides that, for the first time safe detection of a preceding vehicle (i) at time t _k = t _Det , where t _{Det is} the detection time, the initial values from the measured variables are determined as follows:

For the initial distance d0 _j _ _ego , the measured actual distance (dO_ _Le go (tk) = d _meas j (t _k = tDet)), for the own initial driving speed, the measured actual distance Driving speed (v0_ _ego = v _ego _ _me ss (t _k = t _De t)), for the own negative acceleration is the realizable at ABS full braking negative acceleration (a_ _eg o =

ax_min_ABS), for the initial driving speed from the consideration leading vehicle (i) the sum of the measured relative velocity and the measured actual is traveling speed (vO_i = v _reLmessJ (t _k = t _D et) + v _e go_ _mess (tk = t _De t)) and for the negative acceleration of the vehicle in question (i) the sum of the measured relative acceleration and the measured actual acceleration (a_i = arei_messj (tk = t _De t) + a _ego _ _mess (tk = t _Det )) used.

If the viewed fore vehicle (i) is not already at standstill at the time t _k considered, if no measured variables are available, alternatively the negative acceleration from the vehicle ahead (i) considered constant may also be considered from a best-case and worst-case consideration Type of vehicle, eg passenger cars or trucks, and thus the maximum braking capacity of the considered

Vehicle (i).

In addition, in principle, when setting the initial distance dfJj ego, an additional safety or reserve distance d _{res can be} taken into account. This results in a reduced distance. The reduced initial distance or detection distance is then given by do_ _Lego (tk = t _Det) = d _me sj (tk = W - d _res.

The movement prediction and the determination of the parameters relevant to the longitudinal collision assessment are carried out for all vehicles present in the surrounding areas. The evaluation of the model equations and the review on

Collision freedom occurs first for the front clearance FR_V, which is bounded by the present in the current lane first further vehicle F1 forward to clarify whether an imminent collision with the first collision avoidance mode K1 can be prevented. Whether an imminent collision can be prevented by immediate full braking with the negative acceleration (a_ _ego = ax_min_ABS) that can be realized in ABS emergency braking depends on the initial distance or the reduced one

Initial distance and during the ABS full braking relative to the vehicle in front (i), in this case the first further vehicle F1, required braking distance.

Thus, after the start of braking at the time t _k both at the end of the braking and at no predicted time during the braking no Are collision should the predicted dynamic actual distance d_ _e go (k, τ) to said first other vehicle F1 not become negative by the vehicle F. This means for the case in which the first further vehicle F1 nor moved the initial distance during braking dO_i_ _{ego (k} t) is greater than the predicted relative standstill braking distance route Sbr_ss_eg ₀ _i (tk) and at the same time larger than the predicted Dynamic-critical braking distance s _br _ _kr _ _e go_i (tk) must be.

The method further provides that upon first safe detection of a

Vehicle ahead (i), in this case the first further vehicle F1, at the time t _k = t _Det the motion prediction of the own vehicle F with ABS full braking (a_ _ego ⁼ ax_min_ABS) is performed and the following condition for a longitudinal collision freedom.

(13) (d0_i_ego (tk)> Sbr_M_ego_i (tk)) AND (d0_ _{1 e} go (tk)> s _{br kr} _ego_i (tk)) is evaluated.

Conclusion: If the evaluation of the movement prediction with ABS full braking of the own vehicle F, in this case a follower vehicle, satisfies the condition (13) for longitudinal collision freedom with the further first vehicle F1, an imminent collision with the first further vehicle F1 with the first

Collision avoidance mode K1 can be prevented.

On the other hand, if the evaluation of the prediction reveals that the condition (13) for longitudinal collision freedom with the first other vehicle F1 is not satisfied, an impending collision with the first collision avoidance mode K1 can not be prevented even with ABS full braking. In this case, for the considered front clearance FR_V in the instantaneous lane results as free space status information "Status_FR_V - not collision free at EGO full braking (NKF)." It is immediately started with the ABS full braking to mitigate collision damage and It is also checked whether there is a target lane or a target corridor, in which or in the avoidance is possible collision-free with simultaneous full braking.

In order to avoid a collision with the first further vehicle F1 in the current lane during an evasion into a collision-free free space FR of a destination lane, the avoidance procedure must be completed before the first further vehicle F1 is reached. This is particularly in driving situations in which the first further vehicle 1, in particular a vehicle in front (i), in the instantaneous Track can be detected very late, z. B. weather conditions,

due to road geometry, limited sensor range, etc., not always given.

The remaining time or the time interval until the own negative acceleration (a_ _e go) braking the vehicle F touches F1 according to the equations (10) and (11) during a braking operation the first other vehicle, depends on the initial distance, the predicted relative standstill Braking distance and the predicted dynamic critical braking distance.

To check whether it is in the instantaneous track after the time t _k were made starting an ABS emergency braking when additional Dodge still an alternate collision with the first other vehicle F1, in the following is the so-called relevant Berühr-time interval or Collision time interval (T _c _ _eg o_i (tk)) determined and compared with the required evasion time. The relevant contact time interval predicted for the avoidance is the result of the evaluation of the

Motion prediction dependent and is determined for each driving situation.

In the event that in the evaluation of the motion prediction the two

inequalities:

(14)

_<Br s _ ss_ego. AND (d0_ _{1 e} go (t _k )> s _br _ kr_ego. _L (tk)) are satisfied, ie that the initial distance present at the time t _k = t _{Det is} less than the relative standstill braking distance, the relevant result Touch time interval to the vehicle in question from the standstill braking distance due to time interval: T _c _ego_i (t _k ) = T _c _ _s s_ego_i (t _k ).

In the case that in the evaluation of the motion prediction for the two inequalities:

(15) (d0j_ego (t _k)> s _br _ ss_ego. J (t _k)) AND (do_ ₁ _eg ₀ (t _k) <s _br _ kr_ego_1 (tk)) results, which means that the initial distance is less than the Dynamically Critical braking distance is the relevant contact time interval results to the considered Vehicle in front of the dynamic-critical braking distance due to:

Tc_ego_l (tk) = T _C _ .kr_ego. (Tk).

In the case that in the evaluation of the motion prediction for the two inequalities:

(16) (D0_ ₁ _ _ego (t _k) <s _br _ ss_ego. J (t _k)) AND (D0_ ₁ _eg _O (t _k) <s _br _ kr_ego. _L (t) results, which means that the initial distance is smaller than the relative standstill braking distance and less than the dynamic-critical braking distance, the relevant contact time interval results to the first considered further

Vehicle F1 from the minimum of the two time intervals to:

Tc _go W = MIN (T _{c ss} _ _eg 0 _MT _c .

The method also provides that upon first safe detection of a

Vorfahrzeuges (i), here the first further vehicle F1, at the time t _k = t _De t the motion prediction of the own vehicle F with ABS full braking is performed and in the event that the evaluation of the motion prediction results in the condition (13) is not met for longitudinal collision freedom with the vehicle in front (i) and thus an imminent collision with the first collision avoidance mode K1 with ABS full braking of the own vehicle F can not be prevented, the relevant collision time interval T _c _ego_i (t _D et) observed preceding vehicle (i) in the instantaneous lane by evaluating the inequalities (14), (15) and (16) with the initial distance (d0_ _{1 e} go (tk = d _{mess_i} (tk = t _De t) -d _res ) is determined.

In the lateral collision assessment unit 8.2, for a considered time point t _k, according to the provided alternative trajectory f _AW, the avoidance time T _AW _ziei required for evasion into a potential target lane is dependent on the required evasion target lateral offset y _A w_ziei and dependent on one during the evasion maximum allowed lateral acceleration ay _max too

( ^{1 7} ) T _{AW 7Jel} (tk) = K _lri ^■ . ^YAW - ^{7 D} + TR + TAw determined. In this case, K _{t (j} denotes a weighting factor determining the path shape of the avoidance trajectory, T _R _ _Le nk a time delay with which one's own vehicle F steers or _avoids, and T _AW _ _sav a safety time interval, eg to take into account measurement inaccuracies.

The evasive lateral offset y _A w required for avoidance can be:

⁽¹⁸⁾ y _A w) = 0.5 _{t-B: GO} + y, _ _AK () + y are defined _sav.

In this case, yi_ _AK (tk) of the distance of an outer edge of the vehicle ahead (i) at the time t _{k in the} direction of escape relative to the central position M of the driver's own vehicle F and y _{sav is} a lateral safety _distance .

With the time t _k measured lateral distance yi_ _A Kji_mess (tk) to the left outer edge of the first further vehicle F1 arises with (yi_ _AK = yi_A _ii_mess (tk)) from equation (18) a necessary left alternate transverse displacement y _AW ji for the Dodge into the left front space FR_V_li and from Equation (17) a required left dodge time T _AWJj .

Correspondingly, with the measured lateral distance (yi_AK_re_mes _S (tk)) to the right outer edge of the first further vehicle F1 from equation (18), a necessary right alternate lateral displacement y _A w_re for the escape into the front right free space FR_V_re and from equation (17) a right-hand avoidance time T _A w_re required for this purpose. For each potential target lane, the required evasion target lateral offset y _A w_ziei is determined. The determination of the direction of evasion and of the final target lateral offset y _A w_ziei is dependent on the driving situation, depending on the presence of collision-free free spaces FR.

In the event that already at the detection time t _k = t _Det an imminent collision with the first further vehicle F1 in the first collision avoidance mode K1 can not be prevented even with ABS full braking of the own vehicle F, is to determine the escape time at ABS full braking still physically realizable the lateral acceleration (ay _max = ay_max_ABS), which either from Driver safety system, in particular from the ABS, is provided by the driving safety device 13 or is determined in the lateral collision assessment unit 8.2 itself used.

In order that, after the start of the avoidance procedure in a collision-free free space FR of a destination lane at time t _k does not yet come to an avoidance collision with the first further vehicle F1 in the current lane, the available relevant contact time interval (T _c _ _eg o_i (t _k )) may be greater than the evasion time required for avoidance (T _AW _ _Zie i (tk)).

The procedure envisages that upon first safe detection of a

Vehicle ahead (i) at the time t _k = t _Det the motion prediction of the own vehicle F with ABS full braking

ax_min_ABS) is performed, the

Dodge time T _AW _ziei Dependent on the required target lateral offset y _A w_ziei and depending on the maximum possible during full braking with ABS

Transverse acceleration (ay _max = ay_max_ABS) is determined for each potential target track and in the event that the evaluation of the motion prediction results in that

Condition (13) for longitudinal collision freedom with a vehicle in front (i) is not met, and thus an imminent collision with a vehicle in front (i) with the first

Collision avoidance mode K1 with ABS full braking can not be prevented, the following condition for the escape time:

(9) (Tc_ego_i (t _k )> T _{AW target} (t _k )) is evaluated.

D. h.: Does the evaluation of the relevant contact time interval and the avoidance time that the condition (19) for avoidance-longitudinal collision freedom with the vehicle in front (i) in the instantaneous lane in ABS full braking and dodging into a target Track is met, can be a threat of collision with the vehicle in front (i), with the first

Collision avoidance mode K1 with ABS full braking can not be prevented, be prevented with the third collision avoidance mode K3.

Before an avoidance procedure is started at the time t _k , it is, however, in principle checked whether a vehicle can already collide with a vehicle already present in the destination lane or entering the destination lane during and after the avoidance procedure into a destination lane , The method provides that, for the first time safe detection of a preceding vehicle (i) in the current lane and further preceding vehicles (p) in potential destination lanes at the time t _k = t _De t motion prediction for the own vehicle F with ABS full braking

ax_min_ABS) and for each of those vehicles (p) be carried out in potential target tracks and each of the conditions for longitudinal collisions with advance vehicles in accordance with equation (13) (d0_p_ _ego (t _k) <s _{br ss} _ _e GO_P (tk)) and _(p _ego D0_ (tk) ^{<Sbr_kr_ego_p} (t _k)) are evaluated.

The check of a potential target track and the decision whether to switch to the potential left or potential right target track will be explained on the example shown in FIG.

In the driving situation shown, both the left and right neighbor lanes are a potential target lane. The others present in the potential target lanes

Vehicles F2, F3 are located in the front surrounding areas UB_V_li, UB_V_re. Since there are no objects in the rear surrounding areas UB_H, UB_H_li, UB_H_re, in this case it is sufficient to check whether there is a longitudinal collision with a preceding vehicle already present in the destination lane after the avoidance procedure in a destination lane (i) comes.

In the longitudinal collision assessment unit 8.1, in addition to the

Bewegungsprädiktionen for the own vehicle F with ABS full braking and for the vehicle in front (i) in the current lane, here the first further vehicle F1, also Bewegungsprädiktionen for the second further vehicle F2 with the

Initial distance dOJi and for the third further vehicle F3 with the right one

Initial distance d0_re performed and set a target track depending on the result of the evaluations. The initial distances are obtained with measured distances for the first further vehicle F1 in the current lane to dO (t _Det ) = dmess_l (tk ⁼ tDet) "d _r es_1 and for the second and third further vehicle F2, F3 to dO li ( t _Det ) = dmess_2 (tk ⁼ Det) "dres_2 and dO_re (t _De t) = d _meas _3 (tk = t _De t) - d _res _3-

Provides the evaluation for the considered left or right potential target lane that the condition (13) for longitudinal collision freedom with the second further vehicle F2 and the third further vehicle F3 and also the condition (19) for evasive longitudinal Collision freedom with the vehicle ahead (1) in the current lane at ABS Full braking of the own vehicle F and dodge in the considered left or right potential target track are satisfied results for the considered left or

right free space FR_V_ziel of the potential destination lane as free space status information "Status_FR_V_ziel = collision free (KF)".

If the evaluations of the conditions for collision freedom (13) and (9) provide that only the left front clearance FR_V_li alone is collision-free, d. H. Status_FR_V_li = collision free (KF), then the left front clearance FR_V_li is defined as the target clearance (FR_ziel = FR_V_li, Status_FR_ziel = collision free (KF)).

If the evaluations of conditions (13) and (19) provide that only the right front clearance FR_V_re alone is collision free, d. H. Status_FR_V_re = collision free (KF), then the right front free space FR_V_re is defined as the target free space (FR_ziel =

FR_V_re, Status_FR_ziel = Collision free (KF)).

If the evaluations of conditions (13) and (19) provide that both the left front clearance FR_V_li and the right front clearance F_V_re are collision-free (ie, Status_FR_V_li = Collision Free (KF) AND Status_FR_V_re = Collision Free (KF)), then the target -Freiraum selected one free space in which the relevant time-to-touch distance or collision delay: (T _c _ _e goji, T _{e C} go_re) or the dynamically smallest distance (d _krJL ego, d _kr _ _ _{e re} go) is greatest.

Thus, the right front clearance FR_V_re is selected as the target space, where: FR_ziel = FR_V_re when (d _{kr re} _ _e go> d _krJLe go i) and it is the left front

Free space FR_V_li chosen as the target free space, where: FR_ziel = FR_V_li, if (d _kr _ _re _ego

The evaluations of conditions (13) and (19) provide that both the left front clearance FR_V_li and the right front clearance FR_V_re are non-collision free (i.e., Status_FR_V_li = Non Collision Free (NKF) AND Status_FR_V_re = Not

Collision-free (NKF)), there is no collision-free target clearance (FR_ziel = 0, Status_FR_ziel = Non Collision Free (NKF)).

Only if the evaluations of the conditions for collision freedom after the

Conditions (13) and (19) provide a collision - free clearance, takes place for Prevention of impending collision a change from the first

Collision avoidance mode K1 in the third collision avoidance mode K3.

The method provides that even after the first detection of a

Fore vehicle (i) constantly at every real time t _k > t _De t a motion prediction of the own vehicle F with ABS full braking (a _ego = ax_min_ABS) is performed and the condition (13) for longitudinal collision freedom and the condition (19) for

Dodge longitudinal collision freedom is checked with a preceding vehicle (i) in order to prevent an imminent collision, which may arise due to changed driving behavior of the vehicle in front (i) or due to newly occurred events.

FIG. 4 shows a path-time diagram for the predicted vehicle movements of the driving situation presented in FIG. 3 for the specific case that the first further one

Vehicle F1 in the current lane of the own vehicle F, the second further vehicle F2 in the left neighbor lane and the third further vehicle F3 in the right lane lane already at the detection time t _k = t _Det stand still and the own vehicle F with a travel speed v_ _ego above the limit travel speed (v_ego> _ _e g ₀ _g _r enz) approaches the stationary first further vehicle F1 in the current lane.

The abscissa of the path-time diagram is the prediction time τ and the ordinate is the path X.

At the time of detection (t _k = t _Det ), the measured distances d are given by _{s si} (tk ⁼ tüet). d _m ess _ ₂ (t _k = t _De t) and d _MESS _ ₃ (tk = t _Det ) the initial distance to the first further vehicle F1 in the current lane d0_m and the lane-related initial distances to the second and third further vehicles F2, F3 in the neighboring tracks d0_re and dOJi with the relation d0_re>dOJi> d0_m.

The paths traveled by vehicles F, F1 to F3 are designated X _F , Xi, X ₂ and X ₃ . In the example shown in FIG. 4, the braking trajectory according to equation (1c) is used. The path position of the own vehicle F results from the distance traveled in the prediction time τ to X _F (r) = s _br _ego (tk, T) ■ The

Travel position X _S S_EGO is the position at which own vehicle F has come to a standstill. It results from the absolute standstill braking distance. With the initial travel speed vO_ego and the ABS full braking, this results in approximately a parabolic course of the eigenlängsbewegung. The dashed unbalanced longitudinal movement for comparison also has an approximate linear course. At the beginning of a path-time consideration, the distance traveled X _{n of} the first further vehicle F1 in the current track results from the initial distance dO_m. The distance traveled X _{2 of} the second further vehicle F2 in the left-hand lane results from the measured initial distance dOJi and the distance traveled by the third third vehicle F3 in the right-hand lane X ₃ results from the measured initial distance dO_re. The in the

Traffic situation determination unit 8.3 delivers evaluation as

Traffic situation type "traffic jams on motorway", since the other vehicles F1 to F3 are already stopped during the detection.

The evaluation of the movement prediction with ABS full braking of the own

Vehicle F, which in this case represents a follower vehicle, in the illustrated

Driving situation results that both the condition (13) for longitudinal collision freedom with the first further vehicle F1 in the current lane and also the condition (13) for longitudinal collision freedom with the second further vehicle F2 in the left neighbor lane does not meet are as (D0_ _m (t _k) <s _br _ss_ego_i (tk)), and also (do_ _N (t _k) <s _{br ss} _ _ _e go_2 (tk)) is.

The predicted collision times or time intervals up to the collision are T _c _ _eg O_i (t _k) and T ^"c_ego_2 (t _k) - the parallel to the time axis motion lines Xi (), X ₂ (r) intersect the line of movement X _F (τ ) of the own vehicle F to the respective ones

Collision times.

For the front clearance FR_V in the current lane results as free space status information Status_FR_V = "Not collision free at full braking of your own

Vehicle F (NKF) ". For the left front clearance FR_V_li in the left neighbor lane results as free space status information" Non collision free (NKF) ".

The evaluation for the right potential target lane provides that the condition (13) for longitudinal collision freedom with the third further vehicle F3 in the right neighbor lane and also the condition (19) for avoidance longitudinal collision freedom with the first further Vehicle F1 are met in the current lane when dodging in the right neighbor lane. The right front clearance FR_V_re in the right neighbor lane is therefore

Collision free. Thus, the right free space results as the target free space (FR_ziel = FR_V_re, Status_FR_ziel = collision free (KF)). Since the first transition condition A is fulfilled, a transition from the first collision avoidance mode K1 to the third collision avoidance mode K3 takes place for collision avoidance.

The triggering of the avoidance process can be done immediately at the detection time (t _k = t _Det ), since according to equation (10) the contact time interval T _c _ _eg o_i (tDet) to the first further vehicle F1 in the current lane greater than that for dodging required time T _A w_ziei (tüet) is. The time reserve until the last possible activation of the

Dodge process results from the difference of the contact time interval T _c _ _eg o_i (toet) and the required time T _A w_ziei (tDet) - The triggering of the avoidance process at a time (t) after the detection time must therefore in the time interval in the

- t _Del )> 0, ie at the latest to a real one

Time at which, according to equation (10), the constantly calculated contact time interval c_ego_i (tj) for the first further vehicle F1 has reached or falls below the time T _A w_ziei (tj) required for dodge ie at (T _c _ _eg o_i (tj) = T _A w_ziei (tj)).

FIG. 5 shows a further driving situation by way of example with four vehicles F, F1 to F3 and two lanes, and FIG. 6 shows the associated travel-time diagram.

Here is the own vehicle F in the right lane, which here represents the current lane. The first further vehicle F1 in the current lane and also the farther second further vehicle F2 and the closer third more

Vehicle F3 in the left neighbor lane are already at the detection time t _k = t _Det still. The own vehicle F with a driving speed v_ _ego above the

Border speed (v_ _ego > v_e _g0 _ _gr enz) approaches the stationary first further vehicle F1 in the current lane.

With the distances d _{meS_i} (tk = tDet) measured at the detection time (t _k = t _De t), d _m ess_2 (tk = t _De t). d _me ss_3 (tk = t _De t) to the other vehicles F1 to F3 and the distance to the leading edge of the near third further vehicle F3 d _mess _ ₃ vK (tk = t _De t), the initial distances with the relation d0_ ₂ _e arise _g o> d0_ _{1 ego} > d0_ ₃ _ _ego > d0_ _3v «_e _g o- The paths traveled by all vehicles F, F1 to F3 are denoted by X _F , Xi, X ₂ and X ₃ and the The distance traveled by the leading edge of the near third further vehicle F3 is designated by Χ ₃ νκ.

The evaluation carried out in the traffic situation determination unit 8.3 results in the traffic situation type "traffic jams on the highway", since the further vehicles F1 to F3 are already stationary during the detection, between the first further vehicle F1 in the current lane and the leading edge (VK) of the near third Vehicle F3 is a spatial gap of length (d0_ ₁ _ _ego - dO_ _3V K_ego) -

The evaluation of the motion prediction of one's own vehicle F, here the following vehicle, with ABS full braking results in that the condition (13) for longitudinal collision freedom with the first further vehicle F1 in the current lane lane is not met. The predicted collision time until the collision with the first one

Vehicle F1 in the current lane is T _c - _eg o_i (t _k ) - The movement line Xi (r) running parallel to the time axis crosses the movement line X _F (r) of the own vehicle F at this collision time at a first point P1.

For the front clearance FR_V in the current lane results as free space status information Status_FR_V = "Not collision free at full braking of your own

Vehicle F (NKF). "The only free space in the left neighbor lane, particularly the left front lane FR_V_li, is forward through the farther second vehicle F2 and rearward through the forward edge of the near third

Vehicle F3 restricted. The condition (19) for avoidance longitudinal collision freedom with the first further vehicle F1 in the current lane when dodging in the left neighbor lane is extended due to this local restriction. For the avoidance of longitudinal collision freedom with the first further vehicle F1 in the current lane and the third additional vehicle F3 in the destination lane which restricts the escape space, the condition for the avoidance time arises

(20) (Tc_eg ₀ _l (tüet) "Tc_ego_3VK (tDet)] ^> TAW_destination (tDet)) -

Since in the present example the condition according to equation (20) is satisfied, an imminent collision with the first further vehicle F1, with the first

Collision avoidance mode K1 with ABS full braking can not be prevented, be prevented with the third collision avoidance mode K3. However, the avoidance operation must not be triggered immediately at the detection time (t _k = t _De t), since there would be a collision with the near third further vehicle F3 on the left target lane.

The predicted collision time until the collision with the leading edge of the third further vehicle F3 in the target lane is T _c - _eg o_3VK (t _k ) - The triggering of the avoidance process at a point in time (t _j ) after the detection time must therefore take place in the time interval in FIG

[ ^T c_ego ( ^tM ) ~ T _{AW Zicl} (tD _e t)]> ( _t.- t _Del )> T _c _ _egojyK (tD "), that is, at the earliest at the real time, at which the equation according to equation (10) constantly calculated contact time interval to the leading edge of the third further vehicle F3 in the left target track T _{c e} go_3VK (tj) has reached zero and at the latest at a real time in which according to equation (10) the constantly calculated contact time interval T _c _ _eg O_i (t _j) for the first other vehicle F1 required for avoidance time T _A w_zi_i (t _j) has reached or to fall below begins ie when (T _c _ _eg O_i (t _j) = T _AW _ziei (t _j )). Under these restrictive conditions, the left front clearance FR_V_li is collision free in the left neighbor lane.

FIG. 7 shows a further driving situation of the traffic situation type "traffic jam on

Highway "exemplified with 4 vehicles and 2 lanes.

Here is the own vehicle F in the right lane, which the

Instantaneous trace represents. In front of the own vehicle F is in the instantaneous lane a first further vehicle F1. In the left lane is a farther second, moving at a slower speed

Vehicle F2 and a coming from behind third additional vehicle F3, which represents a follower vehicle.

The own vehicle F approaches with a driving speed v_ego above the limit driving speed (v_ _ego > v_ _e go_grenz) to the stationary first further vehicle F1 in the current lane and performs an evasive action, the course of which is hinted over the times t1 to t4 , A first time t1 and a second time t2 are times before the triggering of the avoidance process, a third Time t3 is a time during the dodge, and a fourth time t4 represents a time after dodge.

FIG. 8 shows a path-time diagram for the predicted vehicle movements of the driving situation illustrated in FIG. 7, hereinafter also referred to as case 1 and case 2, and also for two further driving situations, also referred to below as case 3 and case 4, which differ by various measured initial distances, initial travel speeds and initial delays of the third additional vehicle F3 present on the destination lane.

In case 1 and case 2, the third further vehicle F3 respectively comes from behind and is thus a follower vehicle for the own vehicle F. In case 3 and case 4, the third further vehicle F3 is in front of the own vehicle F and thus a preceding vehicle.

The evaluation of the motion prediction of the own vehicle F, here as

Following vehicle, with ABS full braking results that the condition (13) for longitudinal collision freedom with the first further vehicle F1 in the current lane track is not met. For the front clearance FR_V in the current track results as

Free space status information Status_FR_V = "Not collision free with full braking with own vehicle F (NKF)".

In case 1, the prediction that the third additional vehicle F3 is always behind the own vehicle F during braking and evasive as well as at a standstill and there is no for the triggering of the avoidance process

There are restrictions (T _c _ _eg o_3 (t _k )> T _c _e _go _i (t _k )> T _AW _ _Zie i (t _k )). Thus, no collision threatens in this case.

In case 2, the prediction yields that the third further vehicle is initially behind its own vehicle F and at a standstill in front of its own vehicle F. The

Movement line X ₃ (r) crosses the movement line X _F (r) of the own vehicle F at the predicted collision time T _c - _eg o_3 (tDet) at a second point P2. To determine this collision time, the measured distance from the leading edge of your own

Vehicle F to the rear of the approaching from behind on the target track third additional vehicle F3 d _mess _3HK (tk = t _De t) based. For the avoidance longitudinal collision freedom with the first further vehicle F1 in the current lane and the third further restricting the escape space

Vehicle F3 in the destination lane, the condition for the escape time at the considered detection time (t _k = t _Det ) results

(21) [(Tc_ego_1 (tDet) ^_ Tc_ego_3 (Det)] ^> T _A w_Target (tüet)) -

Since, in the present example, the condition according to equation (21) is satisfied, an imminent collision with the third further vehicle F1, which coincides with the first

Collision avoidance mode K1 with ABS full braking can no longer be prevented, can be prevented with the third collision avoidance mode K3, unless the avoidance process is triggered immediately at the considered detection time (t _k = t _De t), but at a time (t _j ) after the detection time in

Time interval [T _c _ _eg "_, (toet) - T _AW _ _{ze /} (tD _M )]> (f, -t _Del )> T _{C eg 3} (tD *).

Under this restrictive condition, the left front clearance FR_V_li in the left neighbor lane is collision free and usable as a target clearance.

In case 3, the prediction yields that the third further vehicle F3 is at the beginning in front of the own vehicle F and behind the own vehicle F at standstill. The movement line Χ ₃ (τ) intersects the movement line X _f (T) of the own vehicle F for the predicted collision time T _c _ _{3 e} go (tDet) in a third point P3. The impending collision can not be prevented in this case by additional evasion, since the time remaining to dodge is smaller than the required evasion time [(T _c _ _ego _i (toet) - _c _3_ego (tDet)] <T _AW _ziei (t _D et)), ie the time gap for avoidance is too small.

An earlier evasion to the left target lane would be possible, but leads after the avoidance process to a rear-end collision, in which the own vehicle F then in the target lane to the predicted collision time T _c _ ₃ _ _eg o (tDet) in the third point P3 on the also braking third additional vehicle F3 ascends.

In case 4, the prediction yields that the third further vehicle F3 is located in front of the own vehicle F at the beginning of the prediction consideration and behind the own vehicle F at standstill. The predicted movement line Χ ₃ (τ) of the third further vehicle _{F 3} crosses the movement line X _F (τ) of the own vehicle F Collision time T _c _3_ego (tDet) at a fourth point P4. A collision before and during the potentially possible evasion is not to be feared, but it comes after dodging in the left-hand target lane to a collision, in which the own vehicle F ascends to the third further vehicle F3.

The driving situation illustrated in FIG. 9 shows the case in which no collision-free free space is available both in the left and in the right neighbor lane and, at the same time, the own vehicle F with negative relative speed is the first further vehicle F1 in the current lane approaches.

The free space check of the neighbor tracks provides with the evaluations of

Conditions (13) and (19) that both the left front clearance FR_V_li and the right front clearance FR_V_re are not collision-free, d. H. Status_FR_VJi = Not collision free (NKF) AND Status_FR_V_re = Not collision free (NKF), so there is no collision free target clearance, ie. h .: FR_ziel = 0, Status_FR_ziel = Not collision free (NKF)).

Since no collision-free clearances are present in this driving situation, the first collision avoidance mode K1 remains activated. In the event that the evaluation of the conditions (13) at the detection time t _k = t _Det shows that at least one collision can be prevented with ABS full braking, the result is the check for longitudinal collision freedom for the front clearance FR_V in the current lane as free space status information Status_FR_V = collision free (KF).

It is determined that Bremstrajektorie or negative acceleration ax_min_Komfort for collision avoidance, with which the braking can be performed as comfortable as possible. Determining the most comfortable braking possible

For example, be carried out by means of a suitable search strategy. The

comfortable negative acceleration a_ _ego = ax_min_Comfort is searched within the limits [ax_min_ABS, ax_min_OG]. Here, ax_min_OG with (ax_min_OG> ax_min_ABS) represents an upper limit for the negative acceleration sought.

Even after the first detection (t _k > t _Det ), there is a constant check for longitudinal collision at each instant t _k , wherein the values measured for the actual distance d _mess (t _k ) and the measured values at time t _{k are the} initial values for the motion prediction the Speeds _(e go_ _m ess (tk) = v vO_i _re i_ _MES s (tk) + _e g ₀ _ _m ess (tk)) can be used, so that, when modified driving situation such. B. at einscherenden vehicles, in

can be reasonably responded to and, if appropriate, to another

Bremstrajektorie is changed.

The driving situation illustrated in FIG. 10 shows the case in which a collision-free free space is available in the left-hand neighbor lane and no collision-free free space in the right-hand neighbor lane, and at the same time the vehicle F is at a vehicle speed V_ego above the limit driving speed (FIG. v_ _eg o> v_ _e rence go_ _g) which slowly, with approximately constant speed, driving the first further F1 vehicle approaches in the instantaneous track.

The evaluation of the motion prediction of one's own vehicle F, here a follower vehicle, with ABS full braking results in the condition (13) for longitudinal collision freedom being met with the first further vehicle F1 in the current lane lane.

For the front clearance FR_V in the current lane, free-space status information results Status_FR_V = "Collision-free with ABS full braking (KF)"

Clearance checking of the neighbor lanes provides, with the evaluations of the conditions (13) and the condition (19) for avoidance longitudinal collision freedom, that only the left front clearance FR_V_li is the only collision-free target clearance (FR_ziel = FR_V_li, Status_FR_ziel = collision free (KF)).

The evaluation carried out in the traffic situation determination unit 8.3 results in the traffic situation type "follow-on drive on motorway", since the own vehicle approaches the first further vehicle F1 in the current lane with a low negative

Relative speed approaches. Since the third transition condition C is satisfied, the collision avoidance is transitioned from the first collision avoidance mode K1 to the second collision avoidance mode K2.

In the second collision avoidance mode K2 now becomes that

Avoiding trajectory for collision avoidance determined, with the avoidance is as comfortable as possible, ie the evasion is feasible with the least possible lateral acceleration ay_max_Komfort. The determination of the most comfortable avoidance trajectory f _AW can

For example, be carried out by means of a suitable search strategy, as Figure 1 1 shows.

FIG. 11 shows a flow chart for a search algorithm for the comfortable lateral acceleration ay_max = ay_max_Komfort. After the start, a query (AF1) for the validity of the objects is first performed. Subsequently, a query (AF2), whether the current time is a detection time t _De t, d. h: tk = t _De t-

The comfortable lateral acceleration ay_max = ay_max_Komfort is searched within the limits [ay_max_UG, ay_max_OG], (ay_max_OG> ay_max_UG) represent ay_max_UG a lower limit and ay_max_OG an upper limit for the desired comfortable lateral acceleration ay_max_comfort, where for example the upper limit is set to a value smaller than or equal to the maximum possible lateral acceleration can be set in a full braking with ABS (ay_max_OG <= ay_max_ABS).

A possible search strategy consists, for example, in the detection time (t _k = t _Det ) starting from the upper limit for the maximum lateral acceleration ayjnaxQ) = ay_max_OG of the maximum lateral acceleration iteratively in each search step (j) by a lateral acceleration difference (Aoy), ie ay_max (j ) = ay_max (j-1) - a, as long as to reduce until either the for a proposed alternative trajectory f _AW to

Dodge required avoidance time T _{AW Zie} i (t _Det , ay_max (j)) the contact time interval or collision time interval with unbraked following drive to the first further

Vehicle F1 in the current lane T _FF , (tk = tDet) falls below. D. h .:

{T _A w_ziei (t _De t, ay_i7iax (j)) <Tc_ _FF _ego_i (toet)} or that the lower limit for the maximum lateral acceleration (ay_max (j) <= ay_max_UG) is reached. This query is denoted by AF3 in FIG.

The comfortable maximum lateral acceleration to be used results in (ay_max = ay_max_comfort = ay_max (j-1)). The triggering of the comfortable avoidance process can take place at the earliest immediately at the detection time (t _k = t _Det ). The time reserve until the last possible activation of the comfort avoidance process results from the difference of the contact time interval T _c _FF_ _eg o_i (tDet) and the required time T _AW _ziei (toet, ay_max_Komfort). The triggering of the avoidance process at a time (tj) after the detection time must therefore in the time interval in the

c_FF_e _g o _j (- ^T AW _Z, CI ( ^tDet - ay_max_ comfort)]> (t, -t _Del )> 0, ie at the latest at a real time, at which the constantly calculated contact time interval _CF F_ego_ (t _j ) to the first further vehicle F1 has reached the time required for comfort Dodge T _AW _ziei (tj, ay_max_Komfort) or to

starts below, ie at (T _c _FF_ego_i (t _j ) = T _{AW Zie} i (t _j , ay_max_Komfort)).

After the detection, the behavior of the first one is constantly in the time until the triggering of the comfort avoidance process, ie for each time point (t _j > t _Det )

Vehicle F1 is monitored in the instantaneous lane and with current measurements the initial distance dO_i_ _eg o (tj) = d _{mess_i} (tj), the driving speeds

_e g ₀ _mess (tj) and rei_ _m ess_i (tj) + v _ego _ _mess (tj) as well as the actual acceleration of the first further vehicle F1 a_-, = a _re i_ _{mess 1} (tj) + a _ego _mess (tj ) is determined for the prediction. In this case, a prediction of the movements is performed and checked for collision freedom with the first further vehicle F1 in the current lane.

Furthermore, both the left neighbor track (destination lane) and the right one

Neighbor track at any time in terms of emerging vehicles, especially einscherender vehicles, and monitors obstacles.

An avoidance procedure is triggered at the latest when the collision can no longer be prevented by ABS emergency braking.

Claims

claims

1. A method at least for reducing the consequences of a collision of a vehicle (F), characterized in that

 is activated from a set of at least three different collision avoidance modes (K1 to K3) at least one collision avoidance mode (K1 to K3), wherein

 when activating a first collision avoidance mode (K1)

 Braking maneuver is performed, taking

 - when activating a second collision avoidance mode (K2)

 Evasive maneuvers are performed, and where

 - When activating a third collision avoidance mode (K3) simultaneously a braking maneuver and an evasive maneuver are performed, and wherein

- The respective collision avoidance mode (K1 to K3) is activated in dependence of a separate vehicle speed (v_ego) of the vehicle (F), a traffic situation and / or an available free space in a vehicle environment.

2. The method according to claim 1,

 characterized in that

 the first collision avoidance mode (K1) is activated when the own driving speed (v_ego) of the vehicle (F) is below a predetermined limit speed.

3. The method according to claim 1 or 2,

 characterized in that

 depending on the traffic situation and the free space the second

Collision avoidance mode (K2) or the third collision avoidance mode (K3) is activated when the own driving speed (v_ego) of the vehicle (F) exceeds a limit speed.

4. The method according to any one of the preceding claims,

 characterized in that

 the third collision avoidance mode (K3) is activated when collision is unavoidable and available clearance exists through the first collision avoidance mode (K1).

5. Method according to one of the preceding claims,

 characterized in that

 after completion of the avoidance maneuver, the first collision avoidance mode (K1) is activated.

6. The method according to any one of the preceding claims,

 characterized in that

 the vehicle environment is divided into surrounding areas, wherein an object recognition and a detection of available free spaces is carried out within each surrounding area.

7. The method according to any one of the preceding claims,

 characterized in that

 in the first collision avoidance operating mode (K1), a braking trajectory with a minimally permitted negative longitudinal acceleration (ax_min) is determined, by means of which a collision can be avoided by the braking maneuver.

8. The method according to any one of the preceding claims,

 characterized in that

 for each free space a collision check is carried out, by means of which free space status information is determined.

9. The method according to claim 8,

 characterized in that

 the collision check taking into account a motion prediction

in the motion prediction, movements of further vehicles (F1 to F3) and hindemis objects on a current lane of the own vehicle (F) as well as on at least one adjacent lane are preconditioned. Device (V) at least to reduce the consequences of a collision of a

Vehicle (F),

characterized in that

from a set of at least three different collision avoidance modes (K1 to K3) at least one collision avoidance mode (K1 to K3) can be activated, wherein

 when activating a first collision avoidance mode (K1)

Braking maneuver is feasible, wherein

 - when activating a second collision avoidance mode (K2)

Evasive maneuver is feasible, and where

 - Upon activation of a third collision avoidance mode (K3) simultaneously a braking maneuver and an evasive maneuver is feasible, and wherein

 - The respective collision avoidance mode (K1 to K3) in response to a separate vehicle speed (v_ego) of the vehicle (F), a traffic situation and an available free space in a vehicle environment can be activated.