DE102015202095A1 - A method for predicting the behavior of at least one preceding vehicle traveling in front of an own vehicle. - Google Patents

Abstract

The invention relates to a method for predicting the behavior of at least one preceding vehicle (220) traveling in front of an own vehicle (210) on a first road (250), comprising the steps of: acquiring road-destination information indicative of possible travel destinations comprising further driving on the first road (250) and on a second road (260) branching from the first road, acquiring vehicle destination information of the preceding vehicle (220), and predicting whether the preceding vehicle (220) will continue on the first road (250) or the second road (260) depending on the road-destination information and the vehicle-destination information.

Description

{TECHNICAL AREA}

The present invention relates to a method for predicting the behavior of at least one preceding vehicle traveling in front of an own vehicle.

{BACKGROUND OF THE INVENTION}

Through driver assistance systems, vehicles driving ahead, which drive in front of an own vehicle, can be monitored. These driver assistance systems can, for example, control or regulate a distance between the own vehicle and the preceding vehicle. For example, the speed of the driver's vehicle may be adjusted by driver assistance systems to maintain a constant distance to the vehicle in front.

In particular, the behavior of preceding vehicles can be predicted by driver assistance systems. For example, it can be estimated whether preceding vehicles from the corresponding lane on which they drive will shear into an own lane on which the own vehicle travels. Such a system is from the EP 2 562 060 B1 known.

However, it may be difficult to accurately predict the behavior of a preceding vehicle, in particular, a possible change of lanes such as boarding on an own lane, especially if there is a turn of the road (eg a highway exit) on which the own vehicle and drive the vehicle in front.

The object of the present invention is to improve the quality of forecasts by predictive systems, especially in such situations.

{SUMMARY}

According to the invention, a method for predicting the behavior of at least one preceding vehicle driving in front of an own vehicle on a first road and a driving assistance system for a vehicle having the features of the independent claims are proposed.

The method according to the invention comprises the steps of acquiring road-destination information comprising possible travel destinations that can be reached by further driving on the first road and driving on a second road that branches off from the first road, Acquiring vehicle target information of the preceding vehicle and predicting whether the preceding vehicle will continue to travel on the first road or on the second road, depending on the road-destination information and on the vehicle-destination information ,

The driving assistance system includes a road-destination information unit configured to acquire road-destination information including travel destinations that can be reached by further driving on a first road on which the own vehicle and vehicle driving a preceding vehicle, and driving on a second road branching from the first road, a vehicle destination information unit configured to acquire vehicle destination information of the preceding vehicle, and a prediction unit in communication with the road-destination information unit and the vehicle-destination information unit, which is configured to predict whether the preceding vehicle will continue to drive on the first road or drive on the second road in response to information exchanged with the road destination information unit and the vehicle destination information unit were. Each of the destination information unit, the admission information unit, and the prediction unit may be configured, for example, as a single electronic control unit (ECU) of the own vehicle or as a common electronic control unit (ECU) of the own vehicle.

Advantages and preferred embodiments of the method according to the invention and the driver assistance system according to the invention will become apparent from the following description in an analogous manner.

In particular, the method may additionally include a step of judging whether a road branches off in front of the own vehicle. When it is detected that a road is branching, the steps of acquiring the road-destination information, acquiring the vehicle-destination information, and predicting on which road the preceding vehicle will travel are performed.

The (first) road on which the own vehicle travels is referred to herein as "private road". In an analogous manner, a lane of the own road on which the Eigenfahrzeug drives, referred to as "own lane".

In order to predict on which road the preceding vehicle will drive, the vehicle destination information and the road destination Information compared or correlated with each other. If the vehicle destination information and the road destination information at least partially coincide, it is concluded that the preceding vehicle will continue to drive on the own road. Similarly, the second road is predicted to be the road on which the preceding vehicle will continue to travel if the corresponding road-destination information for the second road and the vehicle-destination information at least partially coincide.

The main road (first road) and the branching road (second road) may be any kind of road, e.g. As a highway or a highway, a highway, or a highway. For example, the private road may be a highway, and the branching road may be a road (eg, a country road) that leaves the highway. The branching road can also be a connection between two highways. The own road includes in particular a turning lane, over which a vehicle can change from the own road to the branching road.

According to a preferred embodiment of the invention, the step of acquiring vehicle destination information comprises acquiring vehicle information of the preceding vehicle. Such registration information may be related to possible destinations for the preceding vehicle and represents a preferred type of vehicle destination information.

The registration information includes, in particular, a location at which the preceding vehicle is permitted, e.g. For example, a specific country, city, region, province, district, and / or county. Alternatively or additionally, the authorization information in particular includes a residence of the preceding vehicle or the driver of the preceding vehicle. In particular, it may be assumed that the location of the registration of the preceding vehicle corresponds to the residence of the driver.

According to another embodiment, acquiring vehicle-destination information comprises communicating with the at least one preceding vehicle by means of a communication system.

For example, communication systems such as V-2-X systems, i. H. Vehicle-to-vehicle or vehicle-to-vehicle systems may be used to obtain vehicle target information of a preceding vehicle. For example, if a preceding vehicle uses a navigation system into which a particular destination has been entered, that particular destination may be communicated to the subject vehicle via such a V-2-X system.

Advantageously, predicting whether the preceding vehicle will continue to travel on the own road or on the second road will include calculating respective probabilities. Calculating and further processing such probabilities provides robust and reliable computational prediction methods.

By predicting on which road the preceding vehicle will continue to drive, it can be predicted in particular whether the preceding vehicle will change from its lane to the turning lane. If the preceding vehicle is driving on the branching road, it will change from its lane to the turning lane. Otherwise, the vehicle in front will not change to the lane.

Specifically, a first probability is determined whether the preceding vehicle will continue to travel on the road, and a second probability is determined whether the preceding vehicle will travel on the branching road. Obviously, the first probability and the second probability relate to each other in such a way that the first probability results as one minus the second probability and vice versa. The determination of one of these two probabilities is therefore sufficient because the other results directly from it. The certain probability must exceed a specific threshold for the control unit to take action.

The behavior of a plurality of preceding vehicles can be predicted, in particular for each preceding vehicle within an appropriate distance in front of the own vehicle, for. B. within 750 m, 500 m, 250 m, 100 m or 50 m in front of the own vehicle. Alternatively or additionally, the behavior of a predetermined number of vehicles in the vicinity of the own vehicle may be predicted, for example, from the one, the two, the three, the four or the five vehicles, the own vehicle in the own lane and / or closest to adjacent lanes. In particular, vehicle destination information as well as first and second probabilities for each vehicle of this plurality of preceding vehicles may be determined.

According to a preferred embodiment of the invention, it is predicted whether the preceding vehicle will change lanes within or on the first or private road. It is especially predicted whether the vehicle ahead of its lane on the first road will shear on an own lane on the first road if the preceding vehicle does not drive in the own lane. This prediction of whether the preceding vehicle will change lanes is made in consideration of the prediction whether the preceding vehicle will continue to travel on the first road or on the second road which branches off.

In particular, a third probability is determined as to whether the vehicle in front will change lanes, in particular whether the vehicle ahead will crash into the own lane.

A prediction of a shunting on the own lane of a preceding vehicle can therefore be performed more precisely, since it is also taken into account whether the preceding vehicle will continue to drive on the first road or whether it will change roads. False predictions of possible shearing can therefore be reduced. Reference is made to driver assistance systems (also referred to as driver assistance systems) for a prior art own vehicle which use methods to predictively predict future movements of vehicles in the vicinity of the own vehicle. For example, such driver assistance systems predict in advance whether vehicles will make a change of lanes. It is advantageous to retrofit such systems by integrating the possibility of carrying out the methods according to the present invention. As an example of such prior art systems, reference is made to US Pat EP 2 562 060 B1 referenced system.

When predicting a possible lurching of the preceding vehicle, special safety measures can be advantageously carried out, for example, the driver of the own vehicle can be warned by visual and / or acoustic means.

On the one hand, the turning lane may, for example, be on one side of the own lane and the lane of the preceding vehicle may be on the other side of the own lane. If the preceding vehicle is driving on the branching road, the preceding vehicle will shear into the own lane to change to the turning lane. In this case special security measures can be taken.

On the other hand, both the lane of the preceding vehicle and the turn lane can be on the same side of the vehicle's own lane. The turn lane is therefore even further away on this side of the own lane than the lane of the preceding vehicle. If the preceding vehicle is driving on the branching road, the preceding vehicle will most likely not crash into the own lane. Taking this into account, false likelihood forecasts can be made by a driver assistance system, such as that from the EP 2 562 060 B1 known system can be reduced or suppressed, whereby the number of necessary interventions by the driver can be reduced and what leads to a pleasant driving feeling. In addition, the implementation of security measures can be reduced or avoided.

Preferably, the prediction whether the preceding vehicle will crash into the own lane is made when the preceding vehicle is approaching an obstacle in its lane. For example, a second preceding vehicle traveling in front of the preceding vehicle may be regarded as an obstacle, especially when this second preceding vehicle is traveling at a lower speed than the preceding vehicle.

In particular, it can be predicted whether the preceding vehicle will evade the obstacle by changing to the turning lane (because the preceding vehicle will change to the branching road) or by switching to the own lane (because the preceding vehicle continues on the own road will drive). In particular, if the obstacle is a second preceding vehicle which is traveling at a lower speed than the preceding vehicle, it can be estimated whether the preceding vehicle will change from its lane to the lane lane or whether it will overtake the second vehicle ahead. In the latter case, the preceding vehicle will most likely shear into the own lane especially when the preceding vehicle is traveling on a lane adjacent to the own lane.

Predicting a possible shearing on the own lane of a vehicle, which travels on an adjacent lane and approaches an obstacle, can therefore be performed more precisely than in the prior art. False predictions can be reduced and the reliability of these predictions can be increased.

In particular, the method for predicting the behavior of the preceding vehicle includes the step of acquiring branch information that indicates whether there is a branch of the road within a predetermined one Distance in front of the own vehicle there. In particular, it is evaluated whether there is a branch of the road within a predetermined distance of 50 km, 25 km, 10 km, 5 km, 1 km, 750 m, 500 m, 250 m, 100 m or 50 m in front of the own vehicle.

In order to accurately and reliably perform the prediction, road and vehicle target information is acquired, in particular, when the road branches off the main road within this predetermined distance. Therefore, it can be estimated in particular whether the vehicle in front will leave the own road via an upcoming turn lane.

Preferably, the road-destination information is acquired by analyzing information provided by a digital map. Such a digital map may be provided to cooperate with a navigation system, a Global Positioning System (GPS) and / or Simultaneous Localization and Mapping (SLAM). Furthermore, the branch information may preferably also be acquired in this way. The road-destination information unit preferably has such a digital map. The branch information can also be acquired by a camera z. B. recognizes traffic signs or recognized vehicles with trajectories that do not match the course of the road, d. H. the trajectory follows the turn lane, which deviates from the course of the own road, or by direct recognition of the course of the road at a turn (eg turning or entrance lane) by lane mark recognition. The current position of the own vehicle can be determined for example by means of GPS and / or SLAM and / or dead reckoning or a combination thereof. Additionally or alternatively, it is also possible to use mobile network information or similar concepts to determine the current position of the vehicle. Such position information may be associated with map information (for example, provided by the digital map and / or a navigation system) to determine which particular road and direction the home vehicle is traveling. The map information can therefore be evaluated to determine upcoming roads that branch off within the predetermined distance in front of the own vehicle.

In particular, travel destinations within a second predetermined distance around the current position of the own vehicle are provided by the digital map, in particular within a second distance of 500 km, 400 km, 300 km, 200 km, 100 km, 50 km, 25 km, 5 km or 1 km around the own vehicle. Possible travel destinations are determined which can be reached directly by driving on the own road or on the branching road. Furthermore, travel destinations can be determined which can be achieved by a maximum number of changes of the current road, eg. A maximum of two times, five times or ten times. In particular, cities with a certain minimum number of inhabitants are considered as possible destinations

Alternatively or additionally, the road-destination information is obtained by analyzing images of road signs acquired by an image unit. The image unit, for example a camera, can be installed in particular in the own vehicle. The branch information may also preferably be acquired by analyzing these images of street signs. The road-destination information unit preferably comprises such an image unit.

In particular, images of the area in front of the own vehicle can be acquired by the imaging unit and analyzed, for example, by means of algorithms or programs to recognize street signs in general and the writing on these street signs in particular. For example, optical character recognition (OCR) may be used to detect potential travel destinations written on street signs.

According to a preferred embodiment, the vehicle destination information, in particular the registration information, of the preceding vehicle is acquired by analyzing images of a license plate of the preceding vehicle acquired by an imaging unit. This imaging unit (eg, a camera) may be the same imaging unit used to acquire the target information, i. H. the imaging unit can perform multiple optical data acquisition tasks. The image unit may therefore preferably also be integrated in the approval information unit. As explained above, images of the area in front of the own vehicle can be acquired and analyzed to recognize license plates and the writing on these license plates, for example by means of optical character recognition (OCR).

By analyzing an image of a license plate in this way, the specific font on the license plate can be recognized. Usually, this document does not necessarily indicate directly the place of registration or the place of residence of the preceding vehicle or its driver. Usually, the font on license plates includes specific codes (numbers and / or letters) relating to specific locations of registration / residence. Therefore, the particular font on the license plate further analyzed to acquire the registration information. In particular, the particular font is compared to a particular database to extract the approval information from the font on the license plate. Thus, a specific country, city, region, province, district, and / or county are derived from the license plate number as approval information.

Preferably, acquiring the vehicle destination information, in particular the registration information, comprises recognizing the vehicle in front by means of recognition means, e.g. B. by radar and / or LIDAR and / or video camera systems. PMD systems (Photonic Mixer Device) can also be used. The area in front of the own vehicle can be monitored and analyzed, for example, by means of the recognition means. When the preceding vehicle is recognized by the recognition means, an image is acquired from the recognized preceding vehicle by means of an image unit. This image unit is in particular the same image unit (eg camera), which was explained above. This image of the recognized preceding vehicle is analyzed to recognize the license plate of the preceding vehicle and its font, as explained above. According to this embodiment, the preceding vehicle is first recognized, and then the position of the license plate is recognized by an image unit within an image region containing the recognized vehicle. This way of acquiring the registration information is more efficient compared to acquiring images of the area in front of the own vehicle and analyzing the entire acquired area of such images by number plates. Incorrect recognition of license plates (eg, street signs, which are erroneously identified as license plates) can therefore be reduced. Furthermore, this way of acquiring the registration information simplifies the mapping of found number plates and thus the registration information to vehicles, which would otherwise require further processing steps.

According to another preferred embodiment, vehicle-destination information is acquired by communication with the preceding vehicle by means of a communication system. For example, it is conceivable that a communication system used by the preceding vehicle communicates the destination of the preceding vehicle to the own vehicle taking advantage of this information.

Preferably, the communication system comprises a V-2-X system (i.e., vehicle-to-vehicle or vehicle-to-infrastructure). Such systems enable effective communication between different road users.

According to a preferred embodiment, predicting the behavior of the preceding vehicle is performed by judging whether the vehicle destination information coincides with one of the possible travel destinations that are achieved by further driving on the first road and / or driving on the second road can. In particular, it is judged whether the place of registration and / or the location of the preceding vehicle or the driver of the preceding vehicle coincide with one of the possible travel destinations.

If the acquired vehicle destination information, in particular the place of registration and / or the place of residence coincide with one of the possible travel destinations, the corresponding street is predicted as the street on which the preceding vehicle will continue to drive. In particular, such a prediction may be performed as a probabilistic modeling, for example, in this case, an initial probability that a vehicle remains on the own road may be considered as increasing. The amount of the increase may be influenced, for example, by factors such as how safe the acquisition of the vehicle destination information is considered and / or how far the matching destination is located and / or how many changes of the road are needed to reach the destination to reach.

Preferably, it is predicted that the preceding vehicle will travel on the branching road when the acquired vehicle destination information, in particular the registration information, coincides with one of the possible travel destinations that can be reached by driving on the branching road. Especially if the Turning lane and the lane of the preceding vehicle are located on opposite sides of the own lane, it can be predicted in this case that the preceding vehicle will shear on the own lane. If both the turning lane and the lane of the preceding vehicle are located on the same side of the own lane, the lane of the preceding vehicle is unlikely to shear into the own lane.

Preferably, it is predicted that the preceding vehicle will continue to travel on the road if the acquired vehicle destination information, in particular the registration information, matches one of the possible travel destinations that can be reached by further driving on the road. For example, if the turning lane and lane of the preceding vehicle are on the same side of the own vehicle, and if the preceding vehicle is approaching an obstacle, the preceding vehicle is likely to be lashed in the own lane. As explained above, such predictions can be realized on the basis of probabilistic modeling.

As explained above, probabilities can be determined as to whether the preceding vehicle will continue on the road or on the branching road. If the acquired vehicle destination information, in particular the registration information, coincides with the own road or branch road destinations, the corresponding probabilities may be set to a predetermined value (eg, 85%, 90%, 95%, 99 % or 100%). The probabilities can also be represented as continuous values. In particular, it can be considered how the place of registration / residence can be reached through the corresponding roads.

For example, if the location of the license / place of residence can be reached via a first route by driving on the branching road and a second route by driving on the next branching road, these two routes can be compared with each other to obtain the first and second likelihoods determine. For example, if the first route is longer than the second route and / or if it takes longer to reach the location of the license / residence via the first route than via the second route, the second probability may be set to a relatively low value (e.g. B. 45%, 40%, 30%, 25%, 20%, 15%, 10% or 5%). On the other hand, if the first route is the shorter and faster route, the second probability may be set to a comparatively high value (eg 85%, 90%, 95%, 99% or 100%).

However, there is a possibility that the vehicle destination information does not match either one of the possible travel destinations that can be reached by further driving on the own road, nor with any of the possible travel destinations that can be reached by driving on the branch road. This means, in particular, that the place of admission, the place of residence and / or a destination of the vehicle in front are not a possible destination within the second predetermined distance around the current position of the own vehicle. This may be the case, for example, if the preceding vehicle is registered in another country or if the preceding vehicle is a rental car, which is permitted at a location relatively far from the current position of the own vehicle.

On the other hand, if a possible destination, which can be derived from the vehicle destination information, can be achieved both by further driving on the own road and by driving on the branch road, it is expedient to further consider criteria, e.g. B. Which road provides the shorter distance and / or time to reach the possible destination.

In cases where the possible travel path of the preceding vehicle can not be determined with sufficient probability, even taking into account such additional criteria, the final probabilities can be set to predetermined values. For example, the first and second probabilities can both be set to 50%, i. H. it is equally probable that the preceding vehicle remains on the own road or changes to the branching road. The two probabilities can also be set to empirically determined values. For example, since it is usually more likely that a vehicle will stay on a current road than leave the current road, the first probability may be set to a comparatively high value (eg, 85% or 90%) and the second For example, probability may be set to a relatively low value (eg 15% or 10%).

The other criteria that can be considered when estimating probabilities are, on the one hand, road status information such as current traffic information, road surface condition (damaged, etc.), weather conditions (eg icy or dry), traffic rules ( for example, prohibition of vehicles of certain types) or, on the other hand, travel criteria such as distance of the destination on possible itineraries, the time needed to reach the destination, to remain a priori probabilities of vehicles on that road (eg determined by Big Data) or sights along the route. For example, if it is known that there is a congestion on a possible route to a possible destination, it may be considered more likely that the vehicle will use a different route.

Preferably, the prediction of the behavior of the preceding vehicle is used for the control of the own vehicle. In particular, special safety measures can be carried out if it is predicted that the vehicle in front will crash into the vehicle's own lane. The current speed or the maximum speed of the own vehicle can for Example can be reduced as a security measure. Alternatively or additionally, the prediction of the behavior of the preceding vehicle is output visually and / or acoustically. A driver of the driver's vehicle can be warned or informed in particular of a possible collapse of the vehicle ahead, z. By a notification on a faucet or on a head-up display of the own vehicle or by an audible warning or a haptic warning signal (eg, vibration of the steering wheel, vibration of the brake, etc.) or a combination thereof. The determined first and / or second probability may also be displayed, e.g. B. on a fitting or a head-up display.

Advantageously, predicting the behavior of the preceding vehicle is used for a cruise control of the own vehicle. By this Adaptive cruise control (ACC), the distance between the own vehicle and a vehicle, which runs in front of the own vehicle in the own lane, be regulated and automatically adjusted. If it is determined that the preceding vehicle will crash into the own lane, special security measures may be taken to adjust the distance between the own vehicle and the preceding vehicle. The sooner and more accurately this prediction can be made, the earlier and gentler the appropriate safety measure can be taken. Thus, a safe distance to the preceding vehicle can be maintained.

{BRIEF DESCRIPTION OF THE DRAWINGS}

Some preferred embodiments of the invention will now be described in more detail, by way of example only, with reference to the accompanying drawings, in which:

1 schematically shows a preferred embodiment of the method according to the invention as a block diagram.

2 schematically shows a first example of an own vehicle, which runs along a road and performs a preferred embodiment of the method according to the invention.

3 schematically shows a second example of an own vehicle, which runs along a road and performs a preferred embodiment of the method according to the invention.

4 schematically shows a third example of an own vehicle, which runs along a road and performs a preferred embodiment of the method according to the invention.

5 schematically shows an inventive driving assistance system for a vehicle, which is adapted to carry out a preferred embodiment of the method according to the invention.

6 schematically shows a map with possible destinations of an own vehicle, which can be determined in the course of a preferred embodiment of the method according to the invention.

7 schematically shows an image of an area in front of an own vehicle, which can be determined in the course of a preferred embodiment of the method according to the invention.

{DESCRIPTION OF EMBODIMENTS}

1 schematically shows a preferred embodiment of the method according to the invention as a block diagram. This method is executed by a driving assistance system of an own vehicle which runs on a road (herein referred to as first road or own road). The own vehicle travels on a certain lane of the own road (herein referred to as own lane).

In step 110 Branch information is acquired. This branching information indicates whether there is a branch of the road within a predetermined distance of 1 km in front of the own vehicle. A branching road is also referred to herein as a second road.

In step 120 road-destination information is acquired. This road-destination information expresses possible travel destinations that can be reached by further driving on the own road and by driving on the branching road.

In step 130 Vehicle-destination information of a preceding vehicle is acquired. According to the described embodiments, this vehicle destination information is acquired by acquiring vehicle registration information as a specific kind of vehicle destination information. In particular, corresponding registration information can be acquired from a plurality of vehicles in front. In this example, a location of the license to which the preceding vehicle is allowed is acquired as the permission information.

Depending on the acquired branch information, the road-destination information and the registration information a behavior of the preceding vehicle in step 140 predicted.

In step 145 a first probability and a second probability are determined. The first probability indicates whether the preceding vehicle will continue driving on the own road. The second probability indicates whether the preceding vehicle will drive on the branching road. Obviously, the first and second likelihoods are in a fixed relationship since the sum of these two probabilities yields one.

When in step 110 determining that there is no turn of the road within 1 km of the own vehicle, the first probability is set to 100% and the second probability to 0%.

If there is a branch of the road within z. B. 1 km in front of the own vehicle, it is judged whether the registration information with one of the possible travel destinations, which are derived from the road-destination information. In particular, it assesses whether the place of admission is a possible destination on the main road or on the branching road.

If the location of the admission is a possible destination on the own road or the branching road, the corresponding probability is set to a comparatively high value, e.g. 90%. The other probability is set to a comparatively low value, e.g. B. 10%.

If the location of the admission is not one of the possible travel lines, the first and second probabilities are set to predetermined values, e.g. B. to 50% each. The first and second likelihoods are also set to these predetermined values if the location of the registration is a possible destination both on the own road and on the branch road.

Depending on the determined first and second likelihood will step 150 in the course of predicting the behavior of the preceding vehicle. In step 150 a third probability is determined that the preceding vehicle will crash into the own lane. step 150 is of course only performed if the vehicle in front does not drive on the own lane.

step 160 will depend on the results of step 140 carried out. In step 160 For example, safety measures are carried out if it is predicted that the vehicle in front will crash into the vehicle's own lane. The speed of the own vehicle can be reduced, for example, when a shearing is predicted. In addition, the driver of the own vehicle can be visually and / or acoustically notified of such possible shearing

2 schematically shows a first example of an own vehicle 210 which is on a first street 250 (Own road) drives.

The main road 250 may for example be a multi-lane road, z. B. a highway. The own vehicle 210 drives on the own lane 251 the main road 250 , The main road 250 includes two more lanes 251 and 253 , A preceding vehicle 220 drives in front of the own vehicle 210 on the lane 251 ,

It is assumed that a second street 260 within 1 km of your own vehicle 210 branches. The branching road 260 For example, an exit or turn lane of the highway 250 or can also be a connection to another highway.

According to step 120 from 1 possible destinations are determined, which by further driving on the own road 250 and by driving on the branching road 260 can be achieved. According to step 130 from 1 becomes a place of approval of the preceding vehicle 220 certainly.

According to step 140 from 1 be a first probability p ₁ , that the preceding vehicle 220 continue on the main road 250 drive, and a second probability p ₂ that the preceding vehicle 220 on the branching road 260 drive is determined. These probabilities p ₁ and p ₂ are shown schematically in FIG 2 shown.

In addition, other probabilities within the first and second probabilities can be determined. For example, probabilities with respect to the respective lanes 251 . 252 . 253 the street 250 be calculated. Thus, a third probability p ₃ can be determined that the preceding vehicle 220 from his lane 251 on the own lane 252 will collapse. This third probability is in 2 also shown schematically.

If the place of approval of the preceding vehicle 220 for example, a possible destination on the branching road 260 is, the second probability p _{2 is set} to 90% and the first probability p ₁ is set to 10%.

There are no obstacles in front of the vehicle ahead 220 in his lane 251 are and there the vehicle ahead 220 most likely on the branching road 260 drive, the third probability p _{3 can be set} to a comparatively low value.

If the place of approval of the preceding vehicle 220 For example, a possible destination on the main road 250 is, the first probability p _{1 is set} to 90% and the second probability p ₂ is set to 10%. There are no obstacles in front of the vehicle ahead 220 in his lane 251 are the preceding vehicle 220 probably continue on his lane 251 drive. For example, the third probability p ₃ may also be set to a comparatively low value.

3 schematically shows a second example of the own vehicle 210 which runs along the main road 250 drives, analogously 2 , In this example, there is an obstacle 230 in front of the vehicle in front 220 in his lane 251 , This obstacle is, for example, a second vehicle in front, which drives in particular slower than the preceding vehicle 251 ,

If the place of approval of the preceding vehicle 220 for example, a possible destination on the main road 250 is, the first probability p _{1 is set} to 90% and the second probability p ₂ to 10%.

In this case, the preceding vehicle will most likely be the obstacle 230 overtake and will most likely be on the own lane 252 Reeve. Therefore, the third probability p _{3 can} be set to a comparatively high value, for example.

If the place of approval of the preceding vehicle 220 on the other hand a possible destination on the branching road 260 is, becomes the preceding vehicle 220 most likely on the second street 260 change and will not be on the own lane 252 Reeve. Therefore, the second probability p _{2 is set} to 90%, and the first probability p ₁ is set to 10%. The third probability p ₃ can be set to a comparatively low value.

4 schematically shows a third example of the own vehicle 210 which is on the main road 210 drives, analogous to the 2 and 3 , In this example, the vehicle in front drives 220 on the lane 253 ,

If the place of approval of the preceding vehicle 220 a possible destination on the branching road 260 is, the second probability p _{2 is set} to 90% and the first probability p ₁ is set to 10%. Because the vehicle ahead 220 the own lane must cross to the branching road 260 In this case, the third probability p ₃ is set to a high value.

If the place of approval of the preceding vehicle 220 on the other hand, for example, a possible destination on the main road 250 is, the first probability p _{1 is set} to 90% and the second probability p ₂ is set to 10%. Since there are no obstacles in front of the vehicle ahead 220 in his lane 253 own the vehicle in front 220 no increased probability of scoring the own lane, therefore the third likelihood p ₃ remains unchanged.

5 schematically shows a driver assistance system according to the invention 300 for a vehicle, which is adapted to carry out a preferred embodiment of a method according to the invention, as with respect to 1 to 4 explained.

The driver assistance system 300 includes a road-destination information unit 310 , which is set up road-destination information according to step 120 from 1 to acquire. In this example, the road destination information unit is 310 also set up the branch information according to step 110 out 1 to acquire.

The Road Destination Information Unit 310 includes a digital map 312 , This digital map 312 can be optional in a navigation system 331 be integrated and can with a Global Positioning System (GPS) 332 be combined. Additionally or alternatively, the road-destination information unit 310 Use data from a map data provider. By means of this digital map 312 For example, the target information may be acquired in a first manner, as described below 6 is explained.

Alternatively or additionally, the road-destination information unit 310 a camera 333 exhibit. Using such a camera 333 For example, the road destination information may be acquired in a second manner, as discussed below 7 is explained.

The Road Destination Information Unit 310 also includes a computing unit 311 , For example, an electronic control unit (ECU), which is adapted to receive data from the digital map 312 and / or the camera 333 have been determined to process and acquire the road-destination information in response to that data.

The driver assistance system 300 also comprises a vehicle destination information unit, which according to the described embodiment as an admission information unit 320 is provided, which is adapted to admission information according to step 130 out 1 to acquire.

The camera 333 for the Road Destination Information Unit 310 can be used in particular is also in the registration information unit 320 contain. The registration information unit 320 may also according to a preferred embodiment, a radar system 334 include. As an alternative to a radar system, a LIDAR system can also be used. For example, the radar system 334 acquire a rough position of the vehicle, and after this position has been projected onto the image, the camera detects and detects 333 the permission information only in this predetermined coarse position in the image. In case of a fault of the radar system 334 can switch back the camera system and scan the whole picture for approval information. Alternatively or additionally, a communication system (V2X) such as vehicle-to-vehicle or vehicle-to-infrastructure may be used to determine the approval information. Of course, both systems, camera with or without radar / lidar and communication system can be combined.

The Road Destination Information Unit 320 also includes a computing unit 321 For example, an electronic control unit (ECU) that is set up to receive data from the camera 333 and / or the radar system 334 have been collected, further processed and the authorization information acquired on the basis of this data.

The computing units 311 and 321 For example, they can be designed as a common arithmetic unit.

The navigation system 331 , the GPS 332 , the camera 333 and the radar system 334 are especially as sensors 330 formed of the own vehicle and can also be used for other systems of the own vehicle.

The driver assistance system 300 includes a prediction unit 340 in communication with the road destination information unit 310 and the Admissions Information Unit 320 , The prediction unit 340 is set up the probabilities according to step 140 of the 1 to determine. The computing units 311 and 321 and the prediction unit 340 For example, they may be configured as a common arithmetic unit.

The prediction unit 340 is still set up to use other systems 350 of the own vehicle depending on the determined probabilities. The prediction unit 340 For example, the speed of the own vehicle may be controlled by driving a motor controller 351 and / or by braking of the own vehicle or reduce, for example, the driver via a fitting and / or a head-up display 352 and / or an audible signal and / or a haptic signal of the own vehicle inform.

6 schematically shows a map 400 with possible destinations, which by further driving on the own road 250 and on the second street 260 can be achieved, which can be determined in the course of a preferred embodiment of the method according to the invention. In relation to 6 will now explain how road-destination information using the digital map 312 the Road Destination Information Unit 310 can be acquired.

About the GPS 332 becomes a current position of the own vehicle 410 determined (own position 410 ). This particular own position 410 is associated with cat information, which is provided by the navigation system 331 to be provided. In particular, this map information is the digital map showing each road within a distance of 50 km from its own position 410 encompasses around. By correlating the determined eigenposition 410 with the map information can the first road (own road) 250 be identified.

Also, the second street 260 which branches off, be recognized. Thus, the branch information is acquired.

Furthermore, other roads 420 . 430 recognized which with the second street 260 are connected. In this example is the main road 250 , analogous to the 2 to 4 a highway. The branching road 260 is a connection between the highway 250 and two other highways 420 and 430 ,

The map information of the navigation system 331 also contain cities 411 . 412 . 413 , which by further driving on the own road 250 can be achieved. These cities can, for example, by leaving the main road 250 be reached at upcoming exits. These cities 411 . 412 . 413 be considered as possible travel destinations determines which by further driving on the own road 250 can be achieved.

Furthermore, there are cities 421 . 422 . 423 included in the map information, which by leaving the main road 250 over the branching road 260 and by driving further on the highway 420 can be achieved. There are also cities 431 . 432 . 433 which are by driving on the branching road 260 and then driving on the highway 430 can be achieved. Therefore, these cities 421 . 422 . 423 . 431 . 432 and 433 as possible destinations determined by driving on the branching road 260 can be achieved.

7 schematically shows an image of an area in front of an own vehicle, which can be determined in the course of a preferred embodiment of the method according to the invention. In relation to 7 explains how road-destination information and registration information by means of the camera 333 can be acquired.

The camera 333 provides an alternative or additional way to acquire road-destination information. By means of the camera 333 can a picture 500 of the area in front of the own vehicle. In 7 is a situation according to 2 shown. The own vehicle drives on its own lane 252 , The preceding vehicle 220 drives on his lane 251 which are adjacent to the own lane 252 is.

The picture 500 is from the arithmetic unit 311 the Road Destination Information Unit 310 analyzed. For example, a special algorithm can be used to create street signs in the image 500 to recognize.

In this example, a first street sign 510 and a second street sign 520 recognized. The first street sign 510 is the main road 250 assigned, the second street sign 520 is the branching road 260 assigned. fonts 511 and 521 on these street signs 510 respectively. 520 indicate possible destinations, which by driving on the appropriate roads 250 respectively. 260 can be achieved. To the writings 511 and 521 To be able to recognize and to be able to process them as target information, leads the arithmetic unit 311 optical character recognition (OCR) off. V-2-X systems (vehicle-to-vehicle or vehicle-to-infrastructure) are also conceivable solutions, wherein z. B. Street sign information can be acquired by means of V-2-X

In an analogous way, the picture becomes 500 from the arithmetic unit 321 the registration information unit 320 analyzed to acquire the registration information. For example, a specially developed algorithm can be used to license plates 530 in the picture 500 to recognize. To a font 531 on the license plate 530 to recognize leads the arithmetic unit 321 optical character recognition (OCR) off. Pattern matching solutions are also possible. Acquiring the registration information is again conceivable by means of V-2-X. V-2-X systems can also be used to determine specific target information of the preceding vehicle.

This font 531 does not provide the approval information directly. The font 531 includes a specific code (numbers and / or letters) corresponding to specific locations of approval.

The derived font is therefore compared with a database which z. B. license plate codes and the associated places of approval contains. Thus, the place of admission becomes the corresponding recognized font 531 is associated as approval information of the preceding vehicle 220 certainly.

As an alternative to analyzing the entire image 500 and trying number plates throughout the picture 500 You can also identify specific areas of the image 500 to be analyzed. As for 5 explained, the registration information unit 320 also a radar or LIDAR system 334 include. The radar system 334 Can be used to drive the vehicle ahead 220 to recognize and locate. As soon as the vehicle ahead 220 through the radar system 334 localized only needs a special area 540 of the picture 500 around the recognized vehicle 220 be analyzed around. This special area 540 is from the arithmetic unit 321 the registration information unit 320 analyzed to acquire the approval information as explained above.

By first the vehicle ahead 220 is recognized and second only the special area 540 around the recognized vehicle 220 analyzed, license plate number misdetection can be reduced.

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, 0024, 0027]

Claims (19)

Method for predicting the behavior of at least one preceding vehicle ( 220 ), which in front of an own vehicle on a first road ( 250 ), comprising the steps of: acquiring road-destination information including possible travel destinations that are further traveled on the first road ( 250 ) and on a second street ( 260 ), which branches from the first road, can be obtained acquiring vehicle-destination information of the preceding vehicle (FIG. 220 ), and predicting whether the preceding vehicle ( 220 ) on the first street ( 250 ) or on the second street ( 260 ), depending on the road-destination information and the vehicle-destination information. The method of claim 1, wherein acquiring vehicle destination information acquiring vehicle information of the preceding vehicle ( 220 ). The method of any one of the preceding claims, wherein acquiring vehicle destination information communicates with the vehicle in front ( 220 ) by means of a communication system. The method of claim 3, wherein the communication system comprises a V-2-X system. Method according to one of the preceding claims, wherein the predicting whether the preceding vehicle ( 220 ) on the first street ( 250 ) or on the second street ( 260 ), involves calculating corresponding probabilities. Method according to one of the preceding claims, with the following further step: predicting whether the preceding vehicle ( 220 ) Lanes, in particular whether the vehicle in front ( 220 ) from its lane ( 251 ) on the first street ( 250 ) on an own lane ( 252 ) on the first street ( 250 ) when the preceding vehicle ( 220 ) not on the own lane ( 250 ) drives, depending on the prediction, whether the vehicle in front ( 220 ) continue on the first street ( 250 ) or on the second street ( 260 ) will drive. The method of claim 6, wherein the predicting whether the preceding vehicle ( 220 ) Lanes ( 252 ) is performed when the vehicle in front ( 220 ) on its lane ( 251 ) an obstacle ( 230 ) approaches. Method according to one of the preceding claims, with the following further step: acquiring branching information which indicates whether there is a branch ( 260 ) the first street ( 250 ) within a predetermined distance in front of the own vehicle ( 210 ) gives. The method of any one of the preceding claims, wherein acquiring road-destination information comprises analyzing information received from a digital map ( 312 ) to be provided. The method of any one of the preceding claims, wherein acquiring road-to-destination information analyzing images ( 500 ) of street signs ( 510 . 520 ) constituted by an image unit ( 333 ) were acquired. The method of any of claims 2 to 10, wherein acquiring approval information analyzing images ( 500 ) of a license plate ( 530 ) of the preceding vehicle ( 220 ) constituted by an image unit ( 333 ) were acquired. The method of claim 11, wherein acquiring registration information recognizes the preceding vehicle ( 220 ) by means of recognition means ( 334 ), Acquiring an image of the preceding vehicle ( 220 ) by the image unit ( 333 ) and analyzing the image of the detected preceding vehicle ( 220 ). The method of any one of the preceding claims, wherein the predicting comprises judging whether the acquired vehicle destination information, in particular the acquired registration information, matches a possible destination derivable from the street destination information and which further Driving on the first street ( 250 ) and / or driving on the second road ( 260 ) can be achieved. The method of claim 13, wherein it is predicted that the preceding vehicle ( 220 ) continue on the first street ( 250 ), when the acquired vehicle destination information, in particular the acquired registration information, coincide with one of the possible travel destinations derivable from the street destination information and which by further driving on the first street ( 250 ) or it is predicted that the preceding vehicle ( 220 ) on the second street ( 260 ), when the acquired vehicle destination information, in particular the acquired registration information, coincide with one of the possible travel destinations, that of the road destination information are derivable and which by driving on the second road ( 260 ) can be achieved. Method according to one of the preceding claims, wherein a first probability (p ₁ ) that the preceding vehicle ( 220 ) continue on the first street ( 250 ) and a second probability (p ₂ ) that the vehicle in front ( 220 ) on the second street ( 260 ) are set to predetermined values when the acquired vehicle destination information, in particular the acquired registration information, does not match one of the possible travel destinations derivable from the road-destination information that may be generated by further driving the first street ( 250 ) or by driving on the second road ( 260 ), or if the acquired vehicle destination information, in particular the acquired registration information, coincide with one of the possible travel destinations derivable from the road destination information obtained both by further driving on the first street ( 250 ) as well as driving on the second road ( 260 ). Method according to one of the preceding claims, wherein predicting the behavior of the preceding vehicle ( 220 ) for the control of the own vehicle ( 210 ) is used and / or issued visually and / or acoustically and / or haptically, in order to warn the driver of the own vehicle ( 210 ). Method according to one of the preceding claims, wherein predicting the behavior of the preceding vehicle ( 220 ) for an adaptive cruise control of the own vehicle ( 210 ) is used. Driver assistance system ( 300 ) for an own vehicle, comprising a road-destination information unit ( 310 ) configured to acquire road-destination information, which includes travel destinations that can be reached by further driving on a first road on which the own vehicle and a preceding vehicle travel, and by driving on a second road, the from the first road, a vehicle destination information unit ( 320 ) configured to acquire vehicle target information of the preceding vehicle and a prediction unit ( 340 ) in communication with the road-destination information unit ( 310 ) and the vehicle destination information unit ( 320 ) configured to predict whether the preceding vehicle will continue driving on the first road or on the second road, depending on information associated with the road-destination information unit ( 310 ) and the vehicle destination information unit ( 320 ) were exchanged. Driver assistance system ( 300 ) according to claim 18, further adapted to carry out one of the features of the preceding claims 2 to 17.

Images