DE102016106983A1 - Method for detecting a possible lane change maneuver of a target vehicle, control device, driver assistance system and motor vehicle - Google Patents

Abstract

The invention relates to a method for detecting a possible lane change maneuver of a target vehicle (6) in which sensor data are received from at least one sensor (4) of an own vehicle (1), wherein the sensor data is an environment (5) of the own vehicle (1) and the target vehicle (6) describe, based on the sensor data, a plurality of parameters for the detection of the possible lane change maneuver is determined, wherein the plurality of parameters of the target vehicle (6), a relative position between the own vehicle (1) and the target vehicle (6) and / or describe a movement of the target vehicle (1), and the possible lane change maneuver is recognized on the basis of the plurality of parameters, wherein for each of at least two of the parameters a probability value (Pi1, Pi2, Pi3, Po1, Po2) is determined which is a probability for the possible Lane change maneuver describes a total probability (Pi, Po) from the respective probability values (Pi1, Pi2, Pi3, Po1, Po2) is determined and the possible lane change maneuver on the basis of the total probability (Pi, Po) is detected.

Description

The present invention relates to a method for detecting a possible lane change maneuver of a target vehicle in which sensor data is received from at least one sensor of an own vehicle, the sensor data describing an environment of the own vehicle and the target vehicle, based on the sensor data, a plurality of parameters for the detection of the possible Lane change maneuver are determined, wherein the plurality of parameters of the target vehicle, a relative position between the own vehicle and the target vehicle and / or a movement of the target vehicle describe, and the possible lane change maneuver is recognized based on the plurality of parameters. Moreover, the present invention relates to a control device for a driver assistance system of a motor vehicle. Furthermore, the present invention relates to a driver assistance system and a motor vehicle.

Different driver assistance systems are known from the prior art, which support a driver when driving the motor vehicle. For example, Adaptive Cruise Control (ACC) adaptive cruise control systems are known which can control the distance to a vehicle in front. Furthermore, so-called congestion assistants are known, which can autonomously maneuver the motor vehicle at low speeds. In these driver assistance systems, usually a distance to a target vehicle is determined, which is located on an own lane of the own vehicle. In this case, it is necessary that these driver assistance systems can be operated reliably depending on the current traffic situation. This is the case, for example, when a target vehicle is performing a lane change maneuver. For example, the target vehicle may change from the own lane to an adjacent lane. There is also the possibility that the target vehicle changes from an adjacent lane to the own lane. In this case, it would be desirable if such a lane change maneuver can be reliably detected or predicted.

This describes the EP 2 562 060 B1 a method for computer-based prediction of motion patterns of at least one target object, such as a vehicle. In this case, sensor data are generated by at least one sensor which physically detects the surroundings of a carrier vehicle. Further, multiple motion pattern alternatives of the target object are calculated and a prediction of motion patterns of the target object is performed using a context-based prediction step. In doing so, the context-based prediction step uses a set of classifiers, each classifier predicts a probability with which the monitored target performs a motion pattern at a particular time.

It is an object of the present invention to provide a solution as to how a lane change maneuver of a target vehicle can be detected more reliably.

This object is achieved by a method by a control device, by a driver assistance system and by a motor vehicle with the features according to the respective independent claims. Advantageous developments of the present invention are the subject of the dependent claims.

In one embodiment of a method for detecting a possible lane change maneuver of a target vehicle, sensor data are received from at least one sensor of an own vehicle, wherein the sensor data describe an environment of the driver's vehicle and the target vehicle. Preferably, based on the sensor data, a plurality of parameters for the detection of the possible lane change maneuver is determined, wherein the plurality of parameters describe the target vehicle, a relative position between the own vehicle and the target vehicle and / or a movement of the target vehicle. Finally, the possible lane change maneuver is preferably recognized based on the plurality of parameters. Furthermore, in each case at least two of the parameters are each a probability value determined, which describes a probability for the possible lane change maneuver. Furthermore, a total probability is preferably determined from the respective probability values and the possible lane change maneuver is recognized on the basis of the total probability.

An inventive method for detecting a possible lane change maneuver of a target vehicle comprises receiving sensor data from at least one sensor of an own vehicle, wherein the sensor data describe an environment of the own vehicle and the target vehicle. In addition, a plurality of parameters for the detection of the possible lane change maneuver is determined, wherein the plurality of parameters describe the target vehicle, a relative position between the own vehicle and the target vehicle and / or a movement of the target vehicle. Furthermore, the possible lane change maneuver is recognized based on the plurality of parameters. In each case, a probability value is determined for at least two of the parameters, which describes a probability for the possible lane change maneuver. Furthermore, an overall probability is determined from the respective probability values, and the possible lane change maneuver is recognized on the basis of the total probability.

With the help of the method, a future lane change maneuver of the target vehicle should be detected or predicted with the help of the own vehicle. The own vehicle may in particular be a motor vehicle having a driver assistance system. This driver assistance system can in turn comprise one or more sensors with which the surroundings of the own vehicle, in which the target vehicle is located, can be detected. The target vehicle may be a road user who is on the road on which the own vehicle is located. In this case, the target vehicle may be located in the own lane, on which also moves the own vehicle. The target vehicle may also be on a lane adjacent to the own lane. With the at least one sensor of the own vehicle measurements can be carried out continuously in which sensor data are provided. These sensor data can describe not only the target vehicle but also other objects in the vicinity of the own vehicle. Based on the sensor data, a plurality of parameters can be determined. This can be carried out, for example, with a control device of the driver's vehicle or with the aid of an electronic control device of the driver's vehicle. These parameters serve to detect the possible lane change maneuver. These parameters can describe the target vehicle itself. For example, the parameters may describe the spatial dimensions of the target vehicle. Furthermore, the parameters may describe the relative position between the own vehicle and the target vehicle. The parameters may also describe a relative position of the target vehicle to the own lane or to the adjacent lane. Furthermore, the parameters may describe a movement of the target vehicle. The movement of the target vehicle may be, for example, the speed of the motor vehicle. The movement of the target vehicle may also be expressed by a motion model describing the future motion of the target vehicle. On the basis of these parameters it can now be recognized whether a lane change maneuver will be carried out by the target vehicle in the future.

According to the invention, it is now provided that for each of at least two of the parameters a probability value is determined which describes the probability of the possible lane change maneuver. On the basis of the respective parameters, a probability value can be determined, which indicates the probability for a future lane change. It is provided that a respective probability value for at least two parameters is determined. From the respective probability values, a total probability can then be determined. It is provided in particular that a weighted sum of the probability values for determining the total probability is determined. Depending on the overall probability then the possible lane change maneuver of the target vehicle can be detected. Thus, the possible lane change maneuver based on different parameters and their probabilities can be determined. This enables reliable prediction of the lane change maneuver of the target vehicle.

Preferably, the respective probability values are determined for a predetermined time interval. In this case, the time interval can be defined such that it extends from a start time to an end time. The start time can be assigned to a start of the lane change maneuver and the end time to an end of the lane change maneuver. The time interval may also be predetermined and may be, for example, two seconds. Here, it can also be provided that the probability values are determined in each case for respective successive time intervals. This allows a current determination of the probability values and thus a reliable prediction of the lane change maneuver.

In one embodiment, as the possible lane change maneuver a lurching of the target vehicle from an adjacent lane to an own lane of the own vehicle is detected. Here, the case can be considered in which the target vehicle is located on the adjacent lane and changes from this to the own lane of the own vehicle. In this case, in particular the case is considered that the target vehicle is in the direction of travel in front of the own vehicle and after performing the lane change maneuver is on the own lane in front of the own vehicle. If this shearing of the target vehicle is reliably determined, a driver assistance system of the own vehicle, for example a proximity control, can be reliably operated.

In one embodiment, a distance to a vehicle ahead on the own lane and a length of the destination vehicle are determined as parameters. In addition, based on a ratio of the distance to the vehicle and the length of the target vehicle, a first probability value for the shearing of the target vehicle is determined. With the help of at least one Sensors of the own vehicle, a distance to a vehicle can be determined, which is located on the own lane in the direction of travel in front of the own vehicle. This vehicle is a different vehicle from the target vehicle. Further, the length of the target vehicle which is on the adjacent lane is determined. The length of the target vehicle can be determined with the aid of the at least one sensor of the own vehicle. If the length of the target vehicle can not be determined with the aid of the at least one sensor, a default value or a typical value for the length of the target vehicle may be used. Further, the ratio of the distance to the other vehicle and the length of the target vehicle is determined. Thus, it can basically be determined whether the target vehicle can drive into the gap between the own vehicle and the further vehicle. For example, if the gap is not sufficiently large, the likelihood is zero or very low that the target vehicle will collapse. However, if the distance to the preceding vehicle is relatively large, for example, a multiple of the length of the target vehicle, the likelihood of shearing is relatively high. Thus, a first probability value for estimating the shearing of the target vehicle can be determined.

In another embodiment, the parameter is a trajectory having a plurality of prediction points, the trajectory describing a future movement of the target vehicle, and a ratio of a number of the prediction points assigned to the own vehicle lane and a number of the prediction points assigned to the neighboring ones , a second probability value for the shearing of the target vehicle is determined. On the basis of the sensor data, a trajectory or a trajectory can be determined, which describes the future movement of the target vehicle. This trajectory may be determined to include or form multiple prediction points. It can be determined how many of these prediction points are arranged on the own lane and how many of these prediction points are arranged on the adjacent lane. From the ratio of the prediction points on the own lane and the prediction points on the adjacent lane can then be determined the probability of shearing the target vehicle.

In this case, the movement path is preferably determined as a function of a movement model, wherein the movement model describes a change in a speed of the target vehicle during the shearing. The motion model can be determined based on a Kalman filter. This movement model may describe the movement of the motor vehicle and in particular the speed, acceleration and / or jerk of the target vehicle. It can be considered that if a certain speed of the target vehicle is reached, this is usually maintained for a longer period of time. When the target vehicle makes the lane change maneuver, the speed changes. This can be taken into account by increasing the lateral velocity in the motion model. In a further embodiment, a lateral distance of the target vehicle to the own vehicle lane is determined, and a third likelihood value for the lurching of the target vehicle is determined on the basis of the lateral distance. The lateral distance describes in particular the distance of the target vehicle from the own lane perpendicular to a direction of travel of the own lane. In this case, furthermore, a threshold value can be predetermined and it can be checked whether the lateral distance of the target vehicle from the own vehicle lane falls below this predetermined threshold value. For example, such a threshold may be 40 cm. If this threshold is undershot, the likelihood of collapse of the target vehicle is relatively high.

In a further refinement, as the possible lane change maneuver, an exhortation of the target vehicle from the own lane to the adjacent lane is detected. The lane change maneuver may also relate to the situation that the target vehicle is on the own lane and changes from this to the adjacent lane. It is provided in particular that the target vehicle is in the direction of travel in front of the own vehicle, while it performs the lane change maneuver. If such an ejection of the target vehicle can be detected early, the driver assistance system of the driver's vehicle or the adaptive cruise control can be reliably controlled.

Furthermore, it is preferably provided that a lateral distance of the target vehicle to the adjacent lane is determined and a first probability value for the luffing of the target vehicle is determined on the basis of the lateral distance. Again, a corresponding threshold for the lateral distance of the target vehicle to the adjacent lane can be specified, which can be assumed to fall below a relatively high probability of a shearing of the target vehicle. In this way, a first probability value for the skiving of the target vehicle can be determined.

In one embodiment, a lateral velocity of the target vehicle is determined and a second probability value for the skiving is determined by the lateral velocity. It is in particular provided that it is checked whether the target vehicle is performing a cornering and the second probability value for the Ausscheren of the target vehicle is determined depending on whether the target vehicle performs a cornering. In the present case, in addition to the determination of the lateral velocity of the target vehicle, it is checked whether the target vehicle is traveling straight or cornering. When the target vehicle is cornering, it already has a lateral velocity component. This can be taken into account when determining the second probability value for the skiving. Thus, the second probability value can be reliably determined as a function of the current driving situation of the target vehicle.

In this case, provision is made in particular for a rate of rotation of the own vehicle and the lateral speed of the own vehicle to be determined for checking whether the target vehicle is turning. In principle, lane markings can be recognized and derived from this, whether the target vehicle is performing the cornering. Alternatively or additionally, the lateral speed of the target vehicle can be determined. Thus, it can be determined whether the target vehicle is currently cornering. Furthermore, the current rate of rotation of the own vehicle can be determined. This can be done on the basis of appropriate rotation rate sensors of the own vehicle. Thus, for example, it can be checked whether the own vehicle is still driving straight ahead and the preceding vehicle is already driving through a curve or whether both the driver's vehicle and the destination vehicle are driving through the curve or if the driver's vehicle is still cornering and the vehicle is already driving straight ahead.

A control device according to the invention for a driver assistance system of a motor vehicle is designed to carry out a method according to the invention. The control device may in particular be formed by an electronic control unit of the motor vehicle.

A driver assistance system according to the invention comprises a control device according to the invention. The driver assistance system may further include at least one sensor with which the target vehicle and / or the environment of the own vehicle can be detected. The at least one sensor may, for example, be an ultrasound sensor, a radar sensor, a laser scanner, a lidar sensor or a camera. The method according to the invention has the advantage that this can be carried out independently of the type of sensor.

A motor vehicle according to the invention comprises a driver assistance system according to the invention. The motor vehicle is designed in particular as a passenger car. The motor vehicle can be used as a separate vehicle in a method according to the invention.

The preferred embodiments presented with reference to the method according to the invention and their advantages apply correspondingly to the control device according to the invention, the driver assistance system according to the invention and the motor vehicle according to the invention.

Further features of the invention will become apparent from the claims, the figures and the description of the figures. The features and feature combinations mentioned above in the description, as well as the features and combinations of features mentioned below in the description of the figures and / or shown alone in the figures, can be used not only in the respectively specified combination but also in other combinations or in isolation, without the frame to leave the invention. Thus, embodiments of the invention are to be regarded as encompassed and disclosed, which are not explicitly shown and explained in the figures, however, emerge and can be produced by separated combinations of features from the embodiments explained. Embodiments and combinations of features are also to be regarded as disclosed, which thus do not have all the features of an originally formulated independent claim. Moreover, embodiments and combinations of features, in particular by the embodiments set out above, are to be regarded as disclosed, which go beyond or deviate from the combinations of features set out in the back references of the claims.

The invention will now be described with reference to preferred embodiments and with reference to the accompanying drawings.

Showing:

1 an own vehicle according to an embodiment of the present invention in a schematic representation;

2 a traffic situation in which the own vehicle is on an own lane and a target vehicle is on an adjacent lane;

3 the traffic situation according to 2 in a further embodiment;

4 first probability values for intrusion of the target vehicle on the own vehicle lane in dependence on a ratio of a distance between the own vehicle and a preceding vehicle and a length of the target vehicle;

5 a trajectory describing a future movement of the target vehicle;

6 Simulation results for a velocity, an acceleration and a jerk, which were determined by means of a motion model;

7 different trajectories, which were determined on the basis of the movement model;

8th a second probability value for the shearing of the target vehicle in dependence on an arrangement of prediction points of the movement path;

9 another traffic situation in which the target vehicle is at a predetermined lateral distance to the own lane;

10 a third probability value for the shearing of the target vehicle as a function of the lateral distance;

11 another traffic situation in which the target vehicle is at a lateral distance to the adjacent lane at a Ausscheren;

12 a first likelihood value for the escape as a function of the lateral distance;

13 various states that are determined to determine cornering of the target vehicle;

14 a further traffic situation in which several target vehicles make a turn;

15 a second likelihood value for the luffing in response to a lateral speed when driving straight ahead of the target vehicle; and

16 the second likelihood value for the luffing as a function of the lateral speed of the target vehicle when cornering the target vehicle.

In the figures, identical and functionally identical elements are provided with the same reference numerals.

1 shows an own vehicle 1 according to an embodiment of the present invention in a plan view. The own vehicle 1 is designed here as a passenger car. The own vehicle 1 includes a driver assistance system 2 which in turn is a control device 3 includes. The control device 3 For example, by an electronic control unit of the own vehicle 1 be formed. In addition, the driver assistance system includes 2 a sensor 4 with the one environment 5 of the own vehicle 1 can be detected. In particular, the environment may 5 in the direction of travel in front of the own vehicle 1 with the help of the sensor 4 be determined. At the sensor 4 it may be, for example, an ultrasonic sensor, a radar sensor, a lidar sensor, a laser scanner or a camera. It can also be provided that the driver assistance system 2 several sensors 4 having.

With the help of the control device 3 should now be based on sensor data with the sensor 4 be provided, a possible lane change maneuver of a target vehicle 6 be determined. This shows 2 a schematic representation of a traffic situation in which the own vehicle 1 on a own lane 7 located. On a neighboring lane 8th is the target vehicle 6 , On the own lane 7 is also located in the direction of travel in front of the own vehicle 1 another vehicle 9 ,

The lane change maneuver to be recognized may be a shearing of the target vehicle from the adjacent lane 8th on the own lane 7 or to a shunting of the target vehicle from the own lane 7 on the adjacent lane 8th act. To the possible lane change maneuver of the target vehicle 6 To recognize or predict this, a total probability is to be determined, which can range between 0% and 100%. For example, a probability of 0% may describe that the target vehicle is on the current lane 7 . 8th remains. For example, a probability value of 100% may describe that the target vehicle 6 performs a lane change at a predetermined time. It is provided in particular that the probability of the lane change maneuver only for one lane 7 . 8th is done at a time. This can affect both straight-ahead and cornering. To determine the total probability, a plurality of parameters are determined based on the sensor data. For example, these parameters can be the target vehicle 6 , one relative position between the own vehicle 1 and the target vehicle 6 and / or a movement of the target vehicle 6 describe. Individual probabilities or respective probability values P _i1 , P _i2 , P _i3 , P _o1 , P _o2 can then be determined for the respective parameters. From these probability _values P _i1 , P _i2 , P _i3 , P _o1 , P _o2 , a total probability _value P _i , P _o can then be determined.

It is provided in particular that the respective probability _values P _i1 , P _i2 , P _i3 , P _o1 , P _{o2 are determined} in a predetermined time interval, which is defined by a start time and an end time. A transition period results between the start time and the end time, wherein different strategies for determining the probability values P _i1 , P _i2 , P _i3 , P _o1 , P _o2 before, during and after this transitional period can be used. The start time can be defined, for example, by the probability being greater than 15%. The end time may be with a changed position of the target vehicle 6 on the new lane 7 . 8th be related. For example, a trajectory 10 for the target vehicle 6 be determined for a time interval of 2 s. Here it can be checked whether the lane change maneuver has been completed and whether the target vehicle 6 on the new lane 7 . 8th located.

In the present case, initially as a lane change maneuver a shearing of the target vehicle 6 from the neighboring lane 8th on the own lane 7 considered. As parameter here can be a distance a between the own vehicle 1 and the other vehicle 9 and a length l of the target vehicle. This is in 2 illustrated. Based on these quantities, a probability p _i1 can then be determined: p _i1 = ln (α · a / l).

The probability p _i1 is a function dependent on the location and the time. In this case, the parameter α may have a value between 1 and 6. The distance a describes the shortest distance between the own vehicle 1 and the other vehicle 9 , Furthermore, a maximum value for the ratio of the distance a and length l can be specified. This may for example have the value 3. This maximum value can then be assigned to a probability of 100%. Overall, a first probability _value P _i1 for the shearing of the target vehicle 6 be determined as follows:

2 shows a traffic situation in which the distance a between the own vehicle 1 and the other vehicle 9 is greater than the length l of the target vehicle 6 , Here is a lane change from the target vehicle 6 be performed. In comparison shows 3 an example in which the distance a is less than the length l of the target vehicle 6 , Here a lane change is not possible, in which the target vehicle 6 between the own vehicle 1 and the other vehicle 9 is moved.

4 shows the first probability _value P _i1 for the shearing in dependence on the ratio of the distance a to the length l. It can be seen that the probability _value P _i1 for ratios is less than 1 0%, since a lane change is not possible here. The greater the distance a between the own vehicle and the further vehicle 9 is, the higher the likelihood of shearing.

5 shows another traffic situation in which a trajectory 10 indicating the future movement of the target vehicle 6 describes, is determined. The trajectory 10 includes a plurality of prediction points 11 , In this case, a number n _{t of} the prediction points 11 on the adjacent carriageway 8th and a number n _{e of} the prediction points 11 on the own lane 7 be determined. From this, the following function can be derived, which is the ratio of the number n _{e of} the prediction points 11 on the own lane 7 and the number n _{t of} the prediction points 11 on the adjacent lane 8th describes:

The parameter β may have a value between 0 and 2. From the function or the probability p _i2 can then a second probability value P _i2 for the Einscheren of the target vehicle 6 be determined:

wobei gilt y_i+1 = y + D_i + D_{i -1}. With the help of the movement path 10 may be the future movement of the target vehicle 6 be estimated. It takes into account that the dynamic state of the target vehicle 6 will not change significantly if the target vehicle 6 once reached a certain speed. Speed changes and in particular changes in the lateral velocity v _{y occur} , for example, during the lane change maneuver. Therefore, a motion model is used which, in addition to the lateral velocity v _y, also takes into account a lateral acceleration a _y and a jerk J _y in the lateral direction. In addition, an exponential term is provided, which is the drop of the considered lateral velocity v _y during the lane change. In this case, the equation of motion D _i for a step i can be determined according to the following forms:

where is true y _{i + 1} = y + D _i + D _{i -1} .

T describes the time and the constant d the waste. With this movement or with this prediction algorithm, during various iterations or steps i, the current position of the target vehicle 6 be determined. The states are recalculated each cycle and a new prediction for a time horizon or the time interval is determined. Here, it is considered that the vehicle state is determined for a predetermined time interval which is used to determine a number of consecutive steps i for the prediction. The individual steps i can result from the quotient of the time interval and the cycle time. For example, if the time interval is 2 seconds and the cycle time is 40 ms, there are 50 steps.

6 shows different simulation results for the velocity v _y , the acceleration a _y and the jerk J _y for different steps i. Further shows 7 a path s as a function of the stages i for different simulations. The simulations were determined for a speed of 15 km / h. Like from the 6 and 7 can be seen, higher paths are obtained for higher values of the lateral velocity v _y . The greater the lateral acceleration a _y , the faster is the drop in the movement, which indicates that the lane change maneuver is completed faster. Further, the waste can be clearly recognized taking into consideration that the lane change is a transient state of the target vehicle 6 is.

8th shows the second probability _value P _i2 for the shearing in dependence on the ratio n _e / n _t . The probability _value P _i2 thus increases with the number n _{e of} the prediction points 11 on the own lane 7 at.

9 shows another traffic situation in which the target vehicle 6 at a predetermined lateral distance d _y to the own lane 7 located. The lateral distance d _y describes the distance between the nearest point of the target vehicle 6 to the own lane 7 , For this purpose, the following function or probability p _i3 can be defined:

Here, the parameter γ may have a value between 0 and 10. Furthermore, provision may be made for a threshold value to be specified for the lateral distance d _y , which may be, for example, 0.4 m. From this, a third probability _value P _i3 for the shearing can be determined:

10 shows the third probability _value P _i3 for the shearing in dependence on the lateral distance d _y . The limit of 0.4 m can be seen in the graph. It can be seen that the probability value P _i3 of 40% significantly increases from the threshold value of 0.4 m to a probability value P _i3. This means that the probability is already determined in the first 4 cm, which corresponds to a probability value P _i3 of 10%. This results in a larger weighting in this area.

From the individual probability _values P _i1 , P _i2 and P _i3 , a total probability P _i for the shearing P _i can then be determined: P _i = v ₁ · P _i1 + v ₂ · P _i2 + v ₃ · P _i3 .

v ₁ , v ₂ and v ₃ represent the weighting factors. In this case, v ₁ = 1/5, v ₂ = 2/5 and v ₃ = 3/5.

In addition, a probability for a Ausschervorgang be determined, in which the target vehicle 6 a lane change from the own lane 7 on the adjacent lane 8th performs. It is provided in particular that the target vehicle 6 in front of the own vehicle 1 located. This is exemplary in 11 illustrated. Here, the lateral distance d _y between the target vehicle 6 and the adjacent lane 8th certainly. From this, the following function p _o1 can be derived: p _o1 = η - δ · d _y .

The constant η describes the minimum distance of the target vehicle to the adjacent lane 8th for the determination of the probability. In this case, it can be taken into account that the motor vehicle, for example, is centered on the vehicle's own lane 7 located. The constant δ may have a value between 0 and 1. From this a total of a first probability value P ₀₁ can be determined for the skiving:

12 shows the first probability _value P _o1 for the excursion as a function of the lateral distance d _y . Again, a threshold value of 40 cm is taken into account. It can be seen that the first probability _value P _o1 increases linearly with the decreasing lateral distance d _y .

As a further parameter, the lateral velocity v _y can be determined. From this, the following function P _o2 can be derived:

Here μ is a probability parameter which can be specified. From this a total of a second probability _value P _o2 can be determined for the Ausscheren:

It should also be checked whether the target vehicle 6 drive straight or cornering. It takes into account that the dynamic behavior of the target vehicle 6 varies, depending on whether it is cornering or driving straight ahead. A method for determining a presence of cornering of the target vehicle 6 is schematic in 13 shown. In this case, in a step S1, the lateral velocity v _{y of} the target vehicle 6 and the current rate of rotation of the own vehicle 1 certainly. The determination of the lateral velocity v _y and the rate of rotation follows in several successive measuring cycles. It can be considered that at least two target vehicles 6 are detected and whose lateral velocity v _{y is} determined. For the rotation rate of the own vehicle 1 a threshold can be provided. In addition, for the lateral velocity v _{y of} the target vehicle 6 a threshold be provided.

When the rotation rate of the own vehicle 1 is less than the threshold and the lateral velocities v _{y of} the target vehicles 6 greater than the respective threshold value results in a first state S2, in which the own vehicle 1 still driving straight ahead and the target vehicles 6 already make a turn. This is schematically in 14 shown. If the yaw rate of the own vehicle 1 is greater than the threshold and the lateral velocities v _{y of} the target vehicles 6 is greater than the threshold, results in a state S3, which describes that both the own vehicle 1 as well as the target vehicles 6 make a turn. If the yaw rate of the own vehicle 1 is greater than the threshold and the lateral velocity v _{y of} the target vehicles 6 is smaller than the threshold, is the own vehicle 1 still in the curve while the target vehicles 6 go straight again. This describes the state S4. If the yaw rate of the own vehicle 1 is smaller than the threshold value and the lateral velocities v _{y of} the target vehicles 6 are smaller than the threshold, is the own vehicle 1 as well as the target vehicles 6 again on the straight ahead. This describes the state S5.

15 shows the second probability _value P _o2 for the skiving function of the lateral velocity v _y for a straight-ahead travel of the target vehicle 6 , Here, the maximum lateral velocity v _{y is} 2 m / s. In comparison shows 16 the second probability _value P _o2 for the deflection as a function of the lateral velocity v _y for a turning of the target vehicle 6 , Here, the maximum lateral velocity v _{y can be} 4 m / s. This takes into account that the target vehicle 6 when cornering has a lateral velocity component.

Starting from the probability _values P _o1 and P _o2 , a total probability _value P _o for the shunting can be determined. P _o = λ ₁ · P _o1 + λ ₂ · P _o2 .

Λ ₁ and λ _{2 describe} the weighting factors. For example, λ ₁ may be 3/5 and λ ₂ may be 2/5.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• EP 2562060 B1 [0003]

Claims (15)

Method for detecting a possible lane change maneuver of a target vehicle ( 6 ), in which sensor data from at least one sensor ( 4 ) of an own vehicle ( 1 ), the sensor data being an environment ( 5 ) of the own vehicle ( 1 ) and the target vehicle ( 6 ) are determined on the basis of the sensor data, a plurality of parameters for the detection of the possible lane change maneuver, wherein the plurality of parameters of the target vehicle ( 6 ), a relative position between the own vehicle ( 1 ) and the target vehicle ( 6 ) and / or movement of the target vehicle ( 1 ), and the possible lane change maneuver is recognized on the basis of the plurality of parameters, characterized in that a likelihood value (P _i1 , P _i2 , P _i3 , P _o1 , P _o2 ) is determined for at least two of the parameters which determines a probability for describes the possible lane change maneuver, a total probability (P _i , P _o ) from the respective probability _values (P _i1 , P _i2 , P _i3 , P _o1 , P _o2 ) is determined and the possible lane change _maneuver on the basis of the total probability (P _i , P _o ) is recognized. A method according to claim 1, characterized in that the respective probability _values (P _i1 , P _i2 , P _i3 , P _o1 , P _o2 ) are determined for a predetermined time interval. A method according to claim 1 or 2, characterized in that as the possible lane change maneuver a shearing of the target vehicle ( 6 ) from an adjacent lane ( 8th ) on an own lane ( 7 ) of the own vehicle ( 1 ) is recognized. Method according to claim 3, characterized in that as a parameter a distance ( 4 ) to one on the own lane ( 7 ) preceding vehicle ( 9 ) and a length (l) of the target vehicle ( 6 ) and a ratio of the distance (a) to the vehicle ( 9 ) and the length ( 6 ) of the target vehicle ( 6 ) a first probability _value (P _i1 ) for the shearing of the target vehicle ( 6 ) is determined. Method according to Claim 3 or 4, characterized in that a trajectory ( 10 ) with several prediction points ( 11 ), the trajectory ( 10 ) a future movement of the target vehicle ( 6 ) and a ratio of a number (n _e ) of the prediction points ( 11 ), which the own lane ( 7 ) and a number (n _t ) of the prediction points ( 11 ), which of the adjacent lane ( 8th ), a second probability value (P _i2 ) for the shearing of the target vehicle ( 6 ) is determined. Method according to claim 5, characterized in that the trajectory ( 10 ) is determined as a function of a movement model, wherein the movement model is a change of a lateral velocity (v _y ) of the target vehicle ( 6 ) at the Einscheren describes. Method according to one of the preceding claims, characterized in that a lateral distance (d _y ) of the target vehicle ( 6 ) to the own lane ( 7 ) and a third probability value (P _i3 ) for the engagement of the target vehicle ( 6 ) is determined on the basis of the lateral distance (d _y ). Method according to one of the preceding claims, characterized in that, as the possible lane change maneuver, an outrun of the target vehicle ( 6 ) from the own lane ( 7 ) to the adjacent lane ( 8th ) is recognized. A method according to claim 8, characterized in that a lateral distance (d _y ) of the target vehicle ( 6 ) to the adjacent lane ( 8th ) and a first probability _value (P _o1 ) for the outrunning of the target vehicle ( 6 ) is determined on the basis of the lateral distance (d _y ). Method according to claim 8 or 9, characterized in that a lateral speed (v _y ) of the target vehicle ( 6 ) and a second probability value (P _o2 ) for the outrunning of the target vehicle ( 6 ) is determined on the basis of the lateral velocity (v _y ). A method according to claim 10, characterized in that it is checked whether the target vehicle ( 6 ) performs a cornering and the second probability value (P _o2 ) for the Ausscheren of the target vehicle ( 6 ) is determined depending on whether the target vehicle ( 6 ) performs a cornering. A method according to claim 11, characterized in that for checking whether the target vehicle ( 6 ) performs the cornering, a rotation rate of the own vehicle ( 1 ) and the lateral velocity (v _y ) of the target vehicle ( 6 ). Control device ( 3 ) for a driver assistance system ( 2 ) of a motor vehicle ( 1 ), which is designed to carry out a method according to one of the preceding claims. Driver assistance system ( 2 ) for a motor vehicle ( 1 ) with a control device ( 2 ) according to claim 13. Motor vehicle ( 1 ) with a driver assistance system ( 2 ) according to claim 14.

Images