DE102008027590A1 - Driving assistance system for vehicles - Google Patents

Abstract

A Control unit (5) gives a current overall risk function respectively for marking lines, barriers, page boundaries and three-dimensional objects that are present around a vehicle, estimates a temporal change in the position each object and calculates a minimum of the total risk function at the vehicle position for the respective time. A Target function is generated for the time, and a Bypass control amount that minimizes the objective function at the time is calculated as the deflection control amount of the vehicle. Risk functions, provided upon movement of the vehicle to the Umfahrungssteuerbetrag are specified for the respective routes. A final evasive route becomes the risk functions the routes selected, and a steering operation and a braking operation are controlled accordingly.

Description

BACKGROUND OF THE INVENTION

1. Field of the invention

The The present invention relates to a driving assistance system for vehicles, the risks with regard to white Lines or marking lines and three-dimensional objects assesses that are present around a vehicle and for example, with a stereo camera, a one-eyed Camera or a millimeter-wave radar, and controls a steering operation or braking such that the Vehicle can take an optimal route.

2. Description of the relevant State of the art

In In recent years, various technologies have been improved The safety of a vehicle has been developed and used in practice. These technologies create a driving environment in front of the vehicle detected for example with a camera or a laser radar, which is mounted in the vehicle. Based on the data about the driving environment becomes obstacles and a vehicle ahead detected, and there will be an alarm process, an automatic braking process and executed an automatic steering operation.

For example, in a technology as used in the JP-A-2004-110346 discloses an obstacle existing around a vehicle, and the current risk potential of the vehicle with respect to the obstacle is calculated. On the basis of the risk potential, the operation of devices in the vehicle is controlled to cause the driver to perform a driving operation with respect to the movement of the vehicle in the front / rear direction and the right / left direction. The operation of the devices in the vehicle is controlled in only one of the forward / backward direction and the right / left direction.

The disclosed in the aforementioned publication However, control becomes strictly dependent on the current one Risk potential and thus can not be effective To address a risk associated with the movements of the person Vehicle and obstacle changes. In other words Does that mean that even at one time it is considered optimal the roadway often rises in the future.

BRIEF DESCRIPTION OF THE INVENTION

The The present invention is in view of the above Circumstances occur, and an object of the invention is in the indication of a driving support system for Vehicles, the current risk and the future risk taking into account the relative movement between assesses a vehicle and an obstacle and the one performs such control that the vehicle drive in a more natural way along an optimal route and thereby the safety can be improved.

A driving support system for vehicles according to a first aspect of the present invention includes an environment recognizing device for detecting an external environment of a vehicle:
a risk provisioning device for specifying the current risk of an object in the detected external environment;
a risk change predictor for predicting a temporal change in the corrected risk by predicting a time change in a position of the object;
a minimum calculating means for calculating the minimum of the risk at a position of the vehicle at a respective time based on the predicted time change in the risk;
a turn-around control amount calculating means for calculating a turn-around control amount of the vehicle on the basis of at least the minimum; and
an avoidance route determining means for determining a final avoidance route by generating an avoidance route of the vehicle on the basis of the avoidance control amount.

A second aspect of the present invention further includes at least one device of
a steering control means for controlling a steering operation based on the turning amount of the vehicle on the final avoidance route; and
a brake control device for controlling a braking operation based on the risk on the final avoidance route.

According to a third aspect of the present invention, the turning-around control amount calculating means forms a target function at a respective time based on a deviation between a lateral position of the vehicle and the minimum and the turning-around control amount at the time, and calculates a turning-around control amount of the target function as the turning-around control amount of the vehicle to that Time minimized.

According to one Fourth aspect of the present invention calculates the minimum calculator the minimum of risk by partial differentiation in the width direction of the vehicle.

According to one fifth aspect of the present invention then provides the risk provisioning device if the object is a white line or marker line the risk is such that it is approximately of the Center of a lane in the direction of the marker line increases.

According to one sixth aspect of the present invention then provides the risk provisioning device if the object is a three-dimensional object, the risk in a probability distribution.

One Seventh aspect of the present invention further includes a Risk correcting facility for correcting the risk provisioning facility given current risk depending on at least a parameter of relative speed and relative Acceleration between the object and the vehicle.

According to one eighth aspect of the present invention corrects the risk correcting device the current risk given by the risk provisioning facility depending on the relative acceleration between the object and the vehicle, so that the current risk with increasing relative acceleration in a direction in which the object approaches the vehicle increases.

at the driving support system for vehicles according to present invention, it is possible not only the current Risk, but also to specify future risks. This allows control of the vehicle in such a way that this with higher security an optimal Route takes.

BRIEF DESCRIPTION OF THE DRAWINGS

In show the drawings:

1 a schematic diagram for explaining the configuration of a driving support system that is installed in a vehicle;

2 a flowchart for explaining a driving assistance control program;

3 a flowchart in which a continuation of the in 2 shown driving assistance control program is shown;

4 a view for explaining a risk function correction routine;

5 a view for explaining an example of a risk function that is present in front of a vehicle;

6A and 6B Characteristic maps for explaining examples of correction coefficients according to the relative velocity and the relative acceleration; and

7A and 7B Views illustrating examples of a generated evasive route and a detour control amount.

DESCRIPTION OF THE PREFERRED EMBODIMENT

An embodiment of the present invention will be described below with reference to FIGS 1 to 7 described.

As with reference to 1 It can be seen, is a driving support system 2 in a vehicle (a driver's own vehicle) 1 , such as in a motor vehicle, installed. The driving support system 2 primarily has a stereo camera 3 , a stereo image recognition device 4 and a control unit 5 on.

The vehicle 1 is further provided with a vehicle speed sensor 11 for detecting the vehicle speed V, a yaw rate sensor 12 for detecting the yaw rate (dφ / dt) and with a main switch 13 equipped with an ON / OFF signal for a driving assistance control is input. The vehicle speed V is in the stereo image recognition device 4 and in the control unit 5 entered. The control rate (dφ / dt) and the drive assist control ON / OFF signal are input to the control unit 5 entered.

The stereo camera 3 serves as a stereo optical system and includes a pair (right and left) CCD cameras, each of which uses a solid-state image sensor such as a charge-coupled device (CCD). The right and left CCD cameras are mounted in the front portion of a ceiling in the vehicle interior in such a manner as to be interposed with a predetermined spacing. The CCD cameras capture stereo images of outer objects from different viewpoints and input data about the images into the stereo image recognition device.

For example, pictures are taken from the stereo camera 3 in the stereo image recognition device 4 processed in the following way. First, distance information is calculated from the amount of misalignment between the corresponding positions in a pair of stereo images taken by the stereo camera 3 in the forward direction of the vehicle 1 are picked up, and a distance image is generated on the basis of the distance information.

These Image data is subjected to a known grouping and with Compared windows of pre-stored three-dimensional data, such as for example, road shape data, sidewall data, and three-dimensional Object data. As a result of the comparison, data about Marking lines and sidewall data regarding demarcations and curbs extracted along the road extend and three-dimensional objects are used in classes of two-wheeled vehicles, Standard size vehicles, over-length vehicles, Pedestrians, electric pylons and others extracted from three-dimensional objects.

In the above-mentioned detected data, the positions of objects are calculated in a coordinate system in which the position of the vehicle 1 the starting point is the x-axis the forward / backward direction of the vehicle 1 represents and the Y-axis the width direction of the vehicle 1 represents. Specifically, the forward / reverse direction lengths of a two-wheeled vehicle, a standard size vehicle, and an excess length vehicle are estimated in advance, for example, at 3 m, 4.5 m, and 10 m, respectively.

Further, the present central widthwise position of the vehicle is calculated from the central position of the detected width and represented in the form of coordinates (x _obstacle , y _obstacle ). If the length of the vehicle can be accurately detected in the fore / aft direction, for example, by vehicle-to-vehicle communication, the above-described central position can be calculated from the data on the length in the fore / aft direction.

In the three-dimensional object data, the relative velocity Vs with respect to the vehicle becomes 1 based on changes in the distance from the vehicle 1 calculated in the X-axis direction and the Y-axis direction. Using the relative velocity Vs and the vehicle speed V of the vehicle considering the vector size, the velocity in the X-axis direction and the velocity in the Y-axis direction (vx _obstacle , vy _obstacle ) of the three-dimensional object are calculated.

The information thus obtained, that is, data about marker lines, boundary delineation data regarding boundaries and curbs running along the road, and three-dimensional object data (type, distance from the vehicle 1 , central position (x _obstacle , y _obstacle ), speed (vx _obstacle , vy _obstacle ) and relative speed Vs with respect to the vehicle 1 ), are in the control unit 5 entered. This is the stereo camera 3 and the stereo image recognition device 4 provided in the present embodiment as a detection device for detecting the external environment.

The control unit 5 receives the vehicle speed V from the vehicle speed sensor 11 , the yaw rate (dφ / dt) from the yaw rate sensor 12 and marker line data, boundary data about boundaries and curbs running along the road, and three-dimensional object data (type, distance from the vehicle 1 , central position (x _obstacle , y _obstacle ), speed (vx _obstacle , vy _obstacle ) and relative speed Vs with respect to the vehicle 1 ) from the stereo image recognition device 4 ,

On the basis of the input data signals described above, the control unit estimates 5 as a risk function or R _obstacle the current risk for each of the front of the vehicle 1 existing objects, such as marker lines, boundaries, page boundaries and three-dimensional objects, in response to a driving assistance control program, which will be described below.

Here, the current risk R _obstacle for a three-dimensional object is corrected so that it increases with the relative velocity Vs in a direction in which the three-dimensional object is the vehicle 1 is approaching, increasing and thus increasing in relative acceleration (dVs / dt) in the direction in which the three-dimensional object approaches the vehicle 1 approaching, increasing.

On the basis of these risk functions R _line and R _obstacle (corrected values) the current total risk function R is given. Thereafter, a temporal change in the position of each object is predicted at a given total risk function R, and thus a temporal change in the overall risk function R is predicted. On the basis of the temporal change in the total risk function R, minimum values ymin (x, t) in the Y-axis direction on the vehicle at the times are calculated.

Further, objective functions J at the time points are determined by deviations between the lateral positions of the vehicle 1 and the minimum or minima y _min (x, t) and deflection control amounts u (t) determined. A turning-around control amount u (t) that minimizes the objective function J is used as the turning-over control amount u (t) of the vehicle 1 calculated at the time. Risk functions R (t), when moving the vehicle 1 are provided for the respective routes, and a final alternate route R (t) f is selected from the risk functions R (t) of the routes.

On the basis of the bypass control amount u (t) on the final alternate route R (t) f, a control signal is sent to an automatic steering control device 23 which serves as a steering control means to thereby execute a steering control. Further, based on the final alternate route R (t) f, a signal is sent to an automatic brake control device serving as a brake control device 22 outputted to thereby perform a brake control. When the signals to the automatic brake control device 22 and the automatic steering control device 23 they are displayed on a display 21 visually displayed, thereby informing the driver.

In other words, it has the control unit 5 the function of a risk provisioning means, a risk correcting means, a risk change predicting means, a minimum calculating means, a turn-around control amount calculating means and an avoidance-route determining means.

The following is one of the driving support system 2 executed driving support control program with reference to the 2 and 3 described representing flow charts.

First, in a step (hereinafter also referred to as "S" for short) 101 required parameters are read, in particular, mark line data, boundary delineation data regarding boundaries and curbs running along the road, and three-dimensional object data (type, distance from the vehicle 1 , In a central position (x _{obstacle, obstacle} y), velocity (vx _obstacle, vy _obstacle), and the relative velocity Vs with respect to the vehicle 1 ).

In a step S102, the current risk function R _line for marking lines (where curbs and side boundaries are treated the same) is calculated by the following formula (1): R line = K line · y 2 (1).

Here, R _{line represents} a predetermined gain. That is, the current risk function R _line for marker lines is given as a quadratic function representing a central axis in the center of a lane defined by right and left marker lines (where boundaries and page boundaries treated the same), as in 5 is shown.

While the risk function R _{line is} a quadratic function in the present embodiment, it may be any function that allows an increase in the risk from the center of the lane toward the marker lines. For example, the risk function R _{line may} be a fourth degree or sixth degree function.

While in the present embodiment, a boundary and a side boundary are treated the same and are provided with a risk function R second-degree _line , also a function other than the risk function R _line for delineation lines and side boundaries may be provided, thereby taking a higher risk than that Derive risk for marking lines.

For example, if the risk function R _line for the right and left marker lines is specified as a second degree function, the function for the boundaries and page boundaries may be changed to a fourth degree or sixth degree function. Even if a second-degree function is similarly used, the gain K _line can be changed to a higher value.

Moreover, the risk function R _line for marking lines is not limited to a function having a central axis in the center of the lane, and the risk value can be set differently by offsetting the central axis for the left and right lines.

In a step S103, the current risk function R _obstacle for three-dimensional objects (a two-wheeled vehicle, a standard-size vehicle, an over-length vehicle, a pedestrian, an electricity pylon, and other three-dimensional objects) is calculated by the following formula (2). R obstacle = K obstacle * Exp (- ((x obstacle - x) 2 / (2 · σ x obstacle 2 )) - ((y obstacle - y) 2 / (2 · σ y obstacle 2 ))) (2).

The terms have the following meaning: K _obstacle a predetermined gain,
σ x _obstacle a predetermined scattering of the object in the X-axis direction and
y _obstacle a predetermined scattering of the object in the Y-axis direction. These scattering σ x _obstacle and σ y _obstacle can be specified such that they with decreasing recognition accuracy of the stereo camera 3 increase.

Further, the scatters σ x _obstacle and σ y _{obstacle may be} set to be the standard, when the object is a standard size vehicle or an oversized vehicle, that they are high when the vehicle is of a standard size Object is a pedestrian or a two-wheeled vehicle, and that they are small, if the object is another three-dimensional object.

Alternatively, the scatters σ x _obstacle and σ y _{obstacle may be} dependent on the overlap rate in the width direction between the vehicle 1 and the three-dimensional target object. In 5 represent a three-dimensional object A1 and a three-dimensional object A2 examples of current risk functions R _obstacle for three-dimensional objects, which are calculated on the basis of the above-mentioned formula (2).

In a step S104, the current risk function R calculated in the step S103 becomes the _obstacle depending on an in 4 corrected risk function (R _obstacle ) correction routine corrected.

In the risk function (R _obstacle ) correction routine, first, in a step S201, a relative velocity Vs of the three-dimensional target object with respect to the vehicle 1 A first correction gain K _s1 is read with respect to a predetermined map (for example, a Vs-K _s1 map as in FIG 6A shown) in a step S202.

In a step S203, a relative acceleration (dVs / dt) is calculated from the relative velocity Vs. In a step S204, a second correction gain K _s2 with respect to a predetermined map (for example, a (dVs / dt) -K _s2 map as shown in FIG 6B shown).

In a step S204, the risk function R _{obstacle is} corrected and output by the following formula (3): R obstacle = K s1 · K s2 · R obstacle (3). Then the program leaves the routine.

In the in 6A shown map Vs-K _s1 , the first correction gain K _{s1 is set} such that it increases with increasing relative speed Vs. Specifically, let it be assumed that the first correction gain K _s1 is 1.0 when the relative speed Vs is 0, and the target object is at the same speed as the speed of the vehicle 1 moves, while when the relative velocity Vs exceeds 0 and the object is the vehicle 1 approaches, the first correction gain K _s1 exceeds 1.0. Thus, the risk function R _obstacle is corrected to a higher value, as can be seen from the above-described formula (3).

While the driver is feeling a danger with respect to the vehicle 1 approaching obstacle, the driver may not have a strong sense of danger to himself from the vehicle 1 away moving obstacle. In view of this fact, the adjustment described above is made. This setting allows a more natural specification of the risk function R _obstacle .

For the (dVs / dt) -K _s2 map, like the 6B shows, the second correction gain K _{s2 is set} such that it increases with increasing relative acceleration (dVs / dt). More specifically, assuming that the second correction gain K _s2 is 1.0 when the relative acceleration (dVs / dt) is 0 and the target object is not acceleration relative to the vehicle 1 If the relative acceleration (dVs / dt) exceeds 0 and the object slows down or slows down rapidly, then the second correction _gain K _s2 exceeds 1.0. As a result, the risk function S _hindrance is corrected to a higher value, as can be seen from the above-described formula (3).

While the driver is feeling a danger with respect to the vehicle 1 approaching obstacle, as it slows down or decelerates rapidly, the driver does not have a strong sense of danger with respect to an obstacle that is getting faster and moving away from the vehicle 1 moved away. In view of this fact, the adjustment described above is made. This setting allows a more natural specification of the risk function R _obstacle . As with renewed reference to 2 3, in a step S105, the current total risk function R is calculated by the following formula (4): R = R line + R obstacle (4).

In a step S106, a position (x _obstacle (t), y _obstacle (t)) of the three-dimensional object detected after t seconds is estimated by the following formula (5): (x obstacle (t), y obstacle (t)) = (x obstacle + vx obstacle · T, y obstacle + v y obstacle · T) (5).

In a step S107, the position (x _obstacle (t), x _obstacle (t)) of the three-dimensional object detected after t seconds estimated at the step S106 for x and y in the step S105 is calculated Total risk function R substituted to thereby give an overall risk function R (x _obstacle (t), y _obstacle (t)) after t seconds.

In step S108, the total risk function R (x _obstacle (t), y _obstacle (t)) after t seconds calculated in step S107 is partially differentiated in the width direction (y direction). Starting from a point where the detected value is 0, a minimum y _min (x, t) in the width direction (y direction) is calculated. In other words, at the minimum, the following condition is met: .delta..sub.R (x obstacle (t), y obstacle (t)) / δy = 0 (6).

In a step S109, a vehicle position (X (t), Y (t)) after t seconds is estimated by the following formula (7): (X (t), Y (t)) = (V · t, V · ∫sinφ (τ) dτ; integer range 0 ≦ τ ≦ t) (7).

Here, φ (t) represents the yaw rate of the vehicle 1 The yaw rate is given by the following formula (8): φ (t) = (dφ / dt) · t + (1/2) · ((d 2 φ / dt 2 ) + (u (t) / Iz)) · t 2 (8th).

there Iz denotes the yaw moment of inertia, and u (t) represents the above-described Umfahrungssteuerbetrag, as additional yaw moment is used.

In a step S110, the above-described vehicle position (X (t), Y (t)) estimated in step S109 is calculated for the minimum y _min (x, t) calculated in step S108 in the y direction to thereby calculate a minimum y _min (X (t) at the vehicle position X (t).

In a step S111, a target function J is determined based on a deviation between the lateral position Y (t) of the vehicle 1 and the minimum y _min (X (t), t) and the bypass control amount u (t) at each time point. Thereafter, a round trip control amount u (t) which minimizes the objective function J is detected at each time.

Like in the 7A and 7B is assumed that an area in which the vehicle 1 from time 0 (current time) to Δt, referred to as target control range, and the period from time 0 to Δt by dt in 1dt, 2dt, 3dt, ..., mdt, ..., (n-2) dt , (n - 1) dt, and ndt (= Δt) is divided.

For example, during a period from time 0 to 1dt, a target function J0 ^~ 1dt is given by the following formula (9), and a roundabout tax amount u (0) that minimizes the target function J0 ^~ 1dt is determined by a known optimization calculation: J0 ~ 1dt = Wy · (y min (X (1dt), 1dt) - Y (1dt)) 2 + Wu · u (0) 2 (9).

there Wy and Wu denote predetermined weighting values.

For example, during a period from 1dt to 2dt, a target function J1d ^~ 2dt is given by the following formula (10), and a roundabout control amount u (1dt) that minimizes the target function J1dt ^~ 2dt is determined from a known optimization computation: J1dt ~ 2dt = Wy · (y min (X (2dt), 2dt) - Y (2dt)) 2 + Wu · u (1dt) 2 (10).

For example, during a period from 2dt to 3dt, an objective function J2dt ^~ 3dt is given by the following formula (11), and a deflection control amount u (2dt) that minimizes the objective function J2dt ^~ 3dt is determined from a known optimization calculation: J2dt ~ 3 dt = Wy · (y min (X (3dt), 3dt) - Y (3dt)) 2 + Wu · u (2dt) 2 (11).

There if there are two minima at time 3dt, you get two Bypass control amounts u (2dt).

During the periods after time 3dt, similar objective functions are specified and detour control amounts are determined. During a period of (n - 1) dt to ndt, for example, is an objective function J (n - 1) dt ^~ ndt by the following formula (12) is predetermined, and a bypass control amount u ((n - 1) dt) of the Target function J (n - 1) dt ^- ndt minimized, is determined by a known optimization computation: J (n - 1) dt ~ ndt = Wy · (y min (X (ndt), ndt) - Y (ndt)) 2 + Wu · u ((n - 1) dt) 2 (12).

Subsequently, in step S112, a risk function R (t) of each route is determined when the vehicle is moving 1 is provided to the bypass control amount u (t), given by the following formula (13): R (t) = R line + R obstacle (13).

These are R _line and R _{Hin to values provided} with the above expressions (1) and (2) when the vehicle 1 is moved by the deflection control amount u (t). These values are given by the following expressions: R line = K line Y (t) 2 (14) R obstacle = K obstacle * Exp (- (x obstacle (t) - x (t)) 2 / 2 * Ax obstacle 2 )) ((Y obstacle (t) - Y (t)) 2 / 2 · σ y obstacle 2 )) (15).

In a step S113, a final alternate route R (t) f is selected from the risk functions (R (t) of the routes given in step S112 In particular, for each of the routes specified in step S112, the maximum value Rmax is determined Rmax is expressed as follows: Rmax = max (R (t)) (0≤t≤Δt) (16).

A Route where the maximum value Rmax is lowest is called final alternate route R (t) f is selected.

cumulative Risk values Rsum (= ∫R (t) dt; integer range 0 ≤ t ≤ Δt) can be found for the routes, and a route, at the value is the lowest, can be considered final Alternative route R (t) f are selected.

If If only one route is specified in step S112, then Route in the step S113 as the final alternative route R (t) f specified.

At the in 7A 1, a route 1 shown by a solid line and a route 2 shown by a broken line are set in step S112, and one of the routes 1 and 2 where the maximum value Rmax is smaller or the cumulative risk value Rsum is smaller is set in FIG Step S113 is selected as the final alternate route R (t) f. The deflection control amounts u (t) of the routes 1 and 2 are in 7B shown.

In step S114, it is determined whether or not there is an area having a value greater than or equal to a predetermined maximum allowable risk value Rlim (R (t) f ≥ Rlim) in the final alternate route R (t) f. If such a range is not present, a steering control command based on the mileage control amount u (t) of the final avoidance route R (t) f becomes the automatic steering control device in a step S117 23 and the program ends.

If in step S114, it is determined that an area exists is, in which R (t) f ≥ Rlim, in a step S115 a brake start point Xbrake and a brake control time Tbrake on the Based on the earliest time at which R (t) f ≥ Rlim is.

Assuming that the earliest time at which R (t) f ≥ Rlim is Tm, the brake starting point Xbraking is given by the following formula (17). Xbrakes = X (Tm) - Bx (17).

Here, Bx represents a predetermined by a predetermined deceleration G braking distance. The braking distance Bx is given by the following formula (18): Bx = (V 2 / (2 · G)) + Bx0 (18).

there Bx0 sets a predetermined distance up to an obstacle Stop and is for example about 2 m.

Of the Brake start time T braking is started by reverse calculation from the above-described brake start point Xbrake detected.

In a step S116, a brake control command based on the control start point Xbrake and the brake start time Tbrake is sent to the automatic brake control device 22 output.

In a step S117, a steering control command is issued to the automatic steering control apparatus on the basis of the turning control amount u (t) of the final avoidance route R (t) f 23 and the program ends.

As described above, according to the embodiment the present invention, the current overall risk function R for each of the target objects present in front of the vehicle, such as white lines or marking lines, boundaries, page boundaries and three-dimensional objects, given. A temporal change in the overall risk function R is determined by predicting a temporal Change in the position of the target predicted.

Based on the temporal change in the total risk function R, a minimum y _min (x, t) in the y-axis direction is calculated at the position of the vehicle for each time point. An objective function J at the time is determined, and a bypass control amount u (t) that minimizes the objective function J is used as the vehicle's deflection control amount u (t) 1 calculated. Subsequently, a risk function R (t) is generated during movement of the vehicle 1 is provided for the roundabout control amount u (t) for each route.

A final alternate route R (t) f becomes the risk functions R (t) of the routes selected. The steering process is on the Base of the diversion control amount u (t) of the final alternative route R (t) f is controlled, and a braking operation is based on the values the final alternate route R (t) f controlled. For this reason a collision avoidance control can be achieved achieve not only an immediate risk, but also future risks are taken into account.

Given the current overall risk R functions for mark lines, boundaries, side boundaries, and three-dimensional objects that exist in front of the vehicle, the current risks to the target objects are identified as risk functions R _line and R _obstacle . The current risk R _obstacle for a three-dimensional object is corrected such that it increases with increasing relative speed Vs in the direction in which the three-dimensional object is the vehicle 1 approaches, and thus with increasing acceleration (dVs / dt) in the direction in which the three-dimensional object the vehicle 1 approaching, increasing.

For this reason it is possible to use the vehicle 1 to control such that it more naturally takes an optimal route and thereby the safety can be improved, while also the relative movement between the vehicle 1 and the obstacle is exactly considered.

While in the present embodiment both a brake control as well as a steering control based on the final Alternative route R (t) f can be executed also made either only a brake control or a steering control become.

The Brake control used in the present embodiment is for explanation only. Another brake control, in which, for example, the gasoline supply is throttled and at a Automatic transmission is switched to lower gears, can be done in combination.

While in the present embodiment, an external environment based on that of the stereo camera 3 recorded image is detected with a one-eyed camera, a millimeter-wave radar or the like.

While the current overall risk function R is set for each of the marker lines, three-dimensional objects, and the like, in front of the vehicle 1 Furthermore, in the present embodiment, a temporal change in the total risk function R is predicted, the prediction of the overall risk function R and the prediction of the time variation thereof can also be performed for three-dimensional objects laterally beside and rear of the vehicle 1 available.

While in the present embodiment, an escape route during the forward movement of the vehicle 1 This can also be generated during a reverse drive of the vehicle 1 be generated by a back of the vehicle 1 existing environment is evaluated.

In the present embodiment, while the current risk R _obstructs a three-dimensional object depending on the relative velocity Vs and the relative acceleration (dVs / dt) with respect to the vehicle 1 is corrected, a correction can be made depending only on the relative speed Vs or only on the relative acceleration (dVs / dt).

1

vehicle

2

Driving assistance system

3

stereo camera

4

Stereo image recognition device

5

control unit

11

Vehicle speed sensor

12

Yaw rate sensor

13

main switch

21

display

22

automatic Brake control device

23

automatic Steering control device

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• - JP 2004-110346 A [0003]

Claims (9)

A driving support system for vehicles, the driving support system comprising: an environment recognition device ( 4 ) for recognizing an external environment of a vehicle; a risk provision facility ( 5 ) for specifying a current risk for an object in the detected external environment; a risk change predictor for predicting a time change in the corrected risk while predicting a temporal change in the position of the object; a minimum calculation device ( 5 ) for calculating a minimum of the risk at a position of the vehicle at a respective time based on the predicted time change in the risk; a bypass control amount calculating means ( 5 ) for calculating a turning-around control amount of the vehicle based on at least the minimum; and an alternative route determining device ( 5 ) for determining a final alternate route by generating an alternate route of the vehicle based on the bypass control amount. A driving support system according to claim 1, wherein at least one means is provided by: a steering control device ( 23 ) for controlling the steering operation based on the turning amount of the vehicle on the final avoidance route; and a brake control device ( 22 ) for controlling the braking operation based on the risk in the final evasive route. Driving support system according to claim 1 or 2, characterized in that said bypass control amount calculating means (16) 5 ) forms an objective function at a respective time on the basis of a deviation between a lateral position of the vehicle and the minimum and the deflection control amount at the time, and calculates a deflection control amount as the deflection control amount of the vehicle at which the objective function at the time is minimized. Driving support system according to one of the claims 1 to 3, characterized in that the minimum calculating means the minimum of risk through partial differentiation in one Width direction of the vehicle calculated. Driving support system according to one of claims 1 to 4, characterized in that, when the object is a marking line, the risk provisioning device ( 5 ) sets the risk such that it increases approximately from the center of a lane toward the marker line. Driving support system according to one of claims 1 to 5, characterized in that, when the object is a three-dimensional object, the risk provisioning device ( 5 ) specifies the risk in a probability distribution. Driving support system according to one of claims 1 to 6, characterized in that it further comprises a risk correction device ( 5 ) for correcting the risk provisioning facility ( 5 ) given current risk as a function of at least one parameter of a relative speed and a relative acceleration between the object and the vehicle. Driving support system according to claim 7, characterized in that the risk-correcting device ( 5 ) from the risk provisioning institution ( 5 ) corrected current risk in dependence on the relative speed between the object and the vehicle, so that the current risk increases with increasing relative speed in a direction in which the object approaches the vehicle. Driving support system according to claim 7, characterized in that the risk-correcting device ( 5 ) from the risk provisioning institution ( 5 ) corrected current risk in dependence on the relative acceleration between the object and the vehicle, so that the current risk increases with increasing relative acceleration in a direction in which the object approaches the vehicle.