DE102015200059A1 - A driver intention prediction apparatus and method and vehicle comprising such a device - Google Patents

Abstract

A device (10) adapted to predict a driver intention at an intersection, comprising: an intersection situation identification unit (20) arranged to receive intersection situation information adapted to identify the intersection situation based on the received intersection situation information, and is arranged to provide an intersection situation identification output, at least one indicator providing unit (30, 32, 34) arranged thereto to receive at least one sensor signal and configured to provide a driver intent indicator based on the at least one received sensor signal, at least two prediction units (24, 26, 28) each configured to: to provide a driver intention prediction output based on at least one driver intent indicator provided by the at least one indicator B provisioning unit, and a selection unit (20) arranged to select one of the prediction units (24, 26, 28) on the basis of the intersection situation identification output, to select the driver intentions unit. Prediction output of the selected prediction unit as the driver intention prediction result of the device (10).

Description

The present invention relates to a device for predicting the intention of the driver of a vehicle approaching an intersection.

Such a device is for example from the JP 2010-097 335 A known. Based on the position of the vehicle detected by a Global Positioning System (GPS) and map data provided by a map database stored in the vehicle, the known driver intention prediction device determines whether the vehicle is at an intersection approaching or not, and, in the case where the driver fails to operate the turn signal of the vehicle, predicts the driver's intention based on the driver's line of sight, and if necessary, communicates the prediction result to other vehicles in the area surrounding the vehicle.

It is the object of the present invention to provide an improved driver intention prediction apparatus.

This object is achieved by a device which is adapted to predict the driver's intention at an intersection, comprising an intersection situation identification unit which is adapted to receive intersection situation information, is adapted to Identifying an intersection situation based on the received intersection situation information, and configured to provide an intersection situation identification output, at least one indicator providing unit configured to receive at least one sensor signal and configured to provide a driver intent indicator based on the at least one received sensor signal, at least two prediction units, each configured to generate a driver intent prediction output based on the at least one driver Intentions indicator provided by the at least one indicator Provision unit is provided, and a selection unit configured to select one of the prediction units based on the intersection-situation identification output, the driver-intention prediction output of the selected prediction unit to provide the driver's intention prediction result of the device.

The present invention was made on the basis of the fact that the inventor became aware that the behavior of the driver at an intersection depends on the specific situation of the intersection. For example, it is easy to understand that the behavior of the driver will be different depending on whether there are traffic lights at the intersection, whether there are traffic signs at the intersection or if the right of way is decided based on the "right to left" rule , The driver intention prediction apparatus according to the present invention analyzes the intersection situation, identifies the specific intersection situation and, depending on the identification result, selects the most suitable prediction unit for predicting the driver intention in that specific one Crossing situation. Accordingly, the driver intention prediction apparatus according to the present invention provides a more reliable prediction result.

It should be noted that in the context of the present invention, the term "driver intention" may refer to, for example, whether the driver will turn left or right at the intersection or continue to drive straight ahead and / or be the driver Accelerate or decelerate the vehicle or continue at an unchanged speed.

The intersection situation information may include information about the stationary situation of the intersection as such, preferably as viewed from the direction in which the driver's vehicle approaches the intersection. In the context of the present invention, the term "stationary situation of the intersection as such" refers to characteristics of the intersection which do not change over time, for example how many roads meet at the intersection and / or whether they are separate and or combined tracks for the possible directions or not (eg, a track for "left-straight-right" or two tracks, a first track for "left-straight" and a second track for "right" or a first track for "left" and a second track for "straight-ahead" or three tracks, a first track for "left", a second track for "straight ahead" and a third track for "right"), and / or whether there are traffic lights at the intersection or not, and / or whether or not there are separate traffic lights for separate lanes, and / or there are traffic signs at the intersection, and / or what type of street signs on the crossroads ng are provided, and similar information.

Additionally or alternatively, the intersection situation information may include information about the transient situation of the intersection as such, preferably as viewed from the direction in which the driver's vehicle is approaching the intersection. In the context of the present invention, the term "transient situation of the intersection as such" refers to characteristics of the intersection which differ over the Time irrelevant to change the traffic situation at the intersection, z. Whether the traffic lights are in operation or not and / or how the current state of the traffic lights is (red-yellow-green), and the like information.

Additionally or alternatively, the intersection situation information may include information about the driver's vehicle approaching the intersection. For example, the driver's behavior may depend on the type of lane on which the driver's vehicle approaches the intersection.

Additionally or alternatively, the intersection situation information may include information about other moving objects. For example, the behavior of the driver may depend on whether there are other moving objects or not, e.g. As cars, motorcycles, bicycles, pedestrians and the like, and / or from the position and / or direction of movement of the other moving objects and the like.

It should be noted, however, that at least some of this information, eg. For example, the latter information concerning the presence of other moving objects need not be taken into account when the crossing situation is identified. Instead, this information may be used if the driver's intention is predicted, as for each maneuver the driver might possibly intend to make, other moving objects, e.g. For example, a vehicle traveling in a direction intersecting the direction of travel of the driver's vehicle could have priority.

To obtain at least some of the above-mentioned information, the driver intention prediction apparatus may include an interface for obtaining intersection situation information from a vehicle position detection apparatus and / or a map database.

Additionally or alternatively, at least some of the above information may be obtained by evaluating images taken by a camera located in the driver's vehicle. Thus, the driver intention prediction apparatus may further include an interface for obtaining intersection situation information from a pattern recognition apparatus that receives images from a camera.

Additionally or alternatively, at least some of the above-mentioned information may be obtained from the intersection itself if the intersection is equipped with an intelligent intersection technology system.

At least one indicator providing unit may calculate the driver intention indicator based on the at least one received sensor signal.

According to a first example, an indicator providing unit may receive a sensor signal providing information about the head swing angle, accumulating that signal for a predetermined time period or for a predetermined driving distance of the vehicle, and the average head swing angle as one of the driver intention indicators provide. For example, an average head swivel angle α of -15 ° <α <+ 15 ° could be taken as an indication that the driver intends to continue driving straight ahead, with an average head swivel angle α of α <-15 ° could be taken as an indication in that the driver intends to turn right and an average head swing angle α of α> + 15 ° could be taken as an indication that the driver intends to turn left.

The head swivel angle information may be obtained, for example, by evaluating images of the driver using pattern recognition methods taken by an on-board camera of the driver's vehicle. Similarly, an average driver's line of vision may be provided as one of the driver intentional indicators.

• • eine geschätzte Zeit bis zu der Kreuzung, welche auf der Grundlage des Abstands zu der Kreuzung und der momentanen Geschwindigkeit des Fahrzeugs des Fahrers berechnet werden kann,
• • ein Gangwechsel-Vorgang, welcher auf Grundlage der Variation eines Sensor-Signals über die Zeit bestimmt wird, welches Informationen über den momentanen Gang des Fahrzeugs des Fahrers bereitstellt,
• • eine Fahrzeuggeschwindigkeit und eine Geschwindigkeitsänderung, welche auf Grundlage von Informationen über die momentane Geschwindigkeit des Fahrzeugs und/oder die momentane longitudinale Beschleunigung des Fahrzeugs und/oder die momentane Betätigungsmenge des Gaspedals des Fahrzeugs und/oder den momentanen Bremsdruck und/oder den momentanen Gang und/oder die geschätzte lokale Krümmung der Straße, welche zu der Kreuzung führt, und/oder den momentanen Zustand der Ampeln bestimmt sind,
• • das Fahrzeug-Folge-Verhalten, welches auf Grundlage von Informationen über den Abstand zu und/oder die relative Geschwindigkeit in Bezug auf ein vorausfahrendes Fahrzeug bestimmt wird,
• • der Typ von Fahrbahnmarkierung (z. B. ”gestrichelt” oder ”durchgezogen”), welche zum Beispiel durch eine Muster-Erkennungs-Vorrichtung bestimmt werden kann, welche Bilder von einer Kamera empfängt,
• • das Vorliegen einer oder mehrerer benachbarter Spuren, welches auf Grundlage der Informationen über die Position und/oder die relative Geschwindigkeit und/oder den Bewegungszustand eines oder mehrerer Objekte bestimmt wird, welche von einem Radar erfasst werden.

Other examples of driver intentions indicators may be:

• An estimated time to the intersection, which can be calculated based on the distance to the intersection and the current speed of the driver's vehicle,
• A gear change operation that is determined based on the variation of a sensor signal over time that provides information about the current gear of the driver's vehicle,
• A vehicle speed and a speed change, which are based on information about the current speed of the vehicle and / or the current longitudinal acceleration of the vehicle and / or the current actuation amount of the accelerator pedal of the vehicle and / or the instantaneous brake pressure and / or the current gear and / or the estimated local curvature of the road leading to the intersection, and / or the current state of the traffic lights,
• The vehicle following behavior, which is determined based on information about the distance to and / or the relative speed with respect to a preceding vehicle,
• The type of lane marker (eg, "dashed" or "solid") that can be determined, for example, by a pattern recognition device that receives images from a camera,
• The presence of one or more adjacent lanes determined based on the information about the position and / or the relative velocity and / or the state of motion of one or more objects detected by a radar.

It should be noted that one and the same sensor signal may be evaluated according to different evaluation methods to provide two or more different driver intention indicators. For example, the vehicle speed could be evaluated to provide a current value of the vehicle speed and / or a current value of the acceleration / deceleration of the vehicle.

Additionally or alternatively, at least one indicator providing unit may provide the sensor signal as received as the driver intention indicator. For example, the current gear, as obtained from a gearshift (or gearshift) sensor, or the turn signal state, as obtained from a turn signal sensor, may be provided as one of the driver intentional indicators.

The selection unit determines, based on the specific characteristics of the intersection situation identified by the intersection situation identification unit, which of the prediction units is best suited for predicting the driver's intention.

In this regard, it should be noted that a similarity of a plurality of intersection situations may result in the same prediction unit selected for a plurality of different intersection situations. For example, the behavior on a straight priority road at a traffic signal intersection is identical to the behavior at a green traffic light intersection: in all situations, the driver's vehicle has right of way for straight-ahead driving and right turn, giving right of way if it turns left. Therefore, the same prediction unit can be selected for all these situations. Thus, the plurality of possible intersection situations may be divided into a smaller plurality of intersection-situational classes, for each of which a specific prediction unit may be selected.

It should also be noted that for a specific intersection situation, not all driver intentional indicators may be required to provide a reliable prediction result.

According to a first example, the gear change provides no benefit at an intersection without traffic lights, but with a stop sign or right of way sign, or at an intersection without traffic lights and no traffic signs, d. H. at an intersection where the right-to-left rule rules. In either of these situations, slowing down to a speed near zero or even stopping the vehicle will be required accompanied by downshifting into the first or second gear due to the forward drive regulation.

According to a second example, the speed and the speed change provide no benefit at a crossroads without traffic lights but with a stop sign or right of way sign. In this situation, slowing down or even stopping the vehicle will be required due to the right of way regulation.

According to a third example, the vehicle following behavior provides no benefit at an intersection without traffic lights but with a stop sign or right of way sign. In this situation, the behavior of the preceding vehicle will be determined by the right-of-way regulation.

According to a fourth example, the lane marker at a crossroads without traffic lights, but with a stop sign or right of way sign, at an intersection without traffic lights and without traffic signs, ie at an intersection where the "right to left" rule applies, no Advantage ready. In these situations, the Road on which the vehicle approaches the intersection, usually only one lane per direction of travel. In addition, the lane mark at an intersection without traffic lights but with traffic signs approaching the driver's car on a priority road turning right does not provide an advantage because lane mark detection fails for roads with a large curvature.

According to a fifth example, the presence of adjacent lanes at an intersection without traffic lights but with a stop sign or right of way sign, at an intersection without traffic lights and without traffic signs, d. H. at an intersection, where the "right to left" rule applies, no advantage. In these situations, the road on which the vehicle approaches the intersection usually only has one lane per direction of travel. As a result, there is no adjacent track. Moreover, the presence of an adjacent lane at an intersection without traffic lights but with traffic signs approaching the driver's car on a priority road turning right does not. Advantage, since the neighbor track detection by a radar for roads with a large curvature based on missing information about the further course of the road is not reliable.

In general, it would be conceivable to nonetheless provide all available driver intention indicators to all present prediction units, relying on the fact that the irrelevant driver intention indicators are automatically excluded when training the prediction units. However, in order to reduce the complexity of the prediction units, it is preferable to forward to each of the prediction units only the respectively relevant driver intention indicators.

According to a first alternative, the driver intention prediction apparatus according to the present invention may comprise a second selection unit which is adapted to display the intersection situation identification output from the intersection situation identification unit and the driver's intention prediction apparatus. Receive Intentions indicators from the at least two indicator providing units, and is adapted to provide at least one driver Intentions indicator based on the intersection situation identification output to the selected by the first selection unit prediction unit , According to this first alternative, the prediction units have a variable connection with the indicator providing units, and therefore the driver intention prediction apparatus according to the present invention may comprise a reduced number of prediction units, since one and the same prediction Unit can be used for different intersection situations using different sets of driver intention indicators for each of the intersection situations.

However, according to a second alternative, it is also conceivable that each prediction unit is permanently connected to at least one associated indicator providing unit.

According to a specific embodiment, at least one prediction unit may comprise a Bayesian network. Bayesian networks provide a compact representation of required data and can be trained using machine learning techniques based on data obtained in real life. In general, however, any other suitable type of classification device may be used, e.g. For example, a fuzzy logic-based classification device, a support vector machine-based classification device, an artificial neural network-based classification device, to name just a few types of suitable classification devices. Which type of classification device is best suited to be used as a prediction unit for a specific class of crossing situations can be decided, for example, with respect to the compactness of the reproduction of the required data and / or the complexity of their training become.

According to another embodiment, the driver intention prediction apparatus according to the present invention may be configured to output the driver intention prediction result to at least one driving assistance apparatus. A first type of driver assistance device may be an active driver assistance device that actively assists the driver, for example, by activating the turn signal if the driver has failed to activate it, and / or by activating the brakes, if so there is the risk that the driver might disregard the right of way of another vehicle, whereas a second type of driver assistance may be a passive driving assistance device, which passively assists the driver, for example by alerting him to a potentially dangerous situation, e.g. B. by optical and / or acoustic signals to draw attention.

According to another embodiment, the driver intention prediction apparatus according to the present invention may include a triggering apparatus configured to initiate the operation of the driver intention prediction apparatus when a predetermined condition has been met. For example, the operation of the driver intention prediction device may be triggered if the distance to an intersection approaching the driver's vehicle is less than a predetermined distance, e.g. B. less than 20 m. The distance to the intersection may, for example, be determined based on data obtained from a vehicle position detection device and / or a map database and / or by evaluating images taken by a camera located in the driver's vehicle and / or the intersection itself, if the intersection is provided with intelligent intersection technology. However, according to an alternative embodiment, the operation of the driver intention prediction device may be triggered if the time until the intersection approaching the driver's vehicle is less than a predetermined time.

In another aspect, the present invention relates to a method for predicting driver intent at an intersection, comprising the steps of receiving intersection situation information, identifying the intersection situation based on the received intersection situation information, receiving at least one sensor Signal, providing at least one driver intention indicator based on the at least one received sensor signal, selecting one of a plurality of prediction methods based on the intersection situation identification output, predicting the driver intention based on the identified one Intersection situation and the at least one provided driver intention indicator using the selected prediction method.

Other embodiments of the driver intention prediction method according to the invention may be taken from the above description of the driver intention prediction apparatus according to the present invention.

In contrast to the driver intention prediction devices known in the art, the present invention provides a context-based prediction by identifying the crossing situation, the most suitable prediction unit based on the identified one Junction situation is selected and the selected prediction unit is fed with the most appropriate set of driver intention indicators based on the identified intersection situation. This makes it possible to predict the future direction of travel, even if there is no physical change in the trajectory of the vehicle.

In the following, two embodiments of the present invention will be explained in more detail with reference to the accompanying drawings.

1 FIG. 12 is a block diagram of a first embodiment of a driver intention predicting apparatus according to the present invention; FIG.

2 FIG. 12 is a block diagram of a second embodiment of a driver intention prediction apparatus according to the present invention; FIG. and

3 to 11 show a plurality of specific crossing situations,

3 Figure 4 shows a two lane intersection for the driver's vehicle's three possible directions of travel, a first lane for left turn and a second lane for straight ahead or right turn, and with separate traffic lights for the separate lanes, with only the traffic light shown to turn left;

4 shows a three-lane intersection for the driver's vehicle's three possible directions of travel, a first left-turn lane, a second straight-lane drive, and a third right turn lane, and with separate traffic lights for the separate lanes; only the traffic light for turning left is shown;

5 shows a three-lane intersection for the driver's vehicle's three possible directions of travel, a first left turn lane, a second straight drive lane and a third right turn lane, and a common traffic light for all lanes; the green light is on;

6 shows a junction with a combined lane for the three possible directions of travel of the driver's vehicle and with a common traffic light for all lanes;

7 Fig. 12 shows an intersection without traffic lights but with traffic signs, a general convenience sign or (indicated by a slash) a right-to-left intersection sign for the direction of travel of the driver's vehicle;

8th shows an intersection without traffic lights, but with traffic signs, namely a stop sign or (indicated by a slash) a Vorfahrgewähren sign for the direction of travel of the driver's vehicle;

9 shows an intersection at which the right of way is decided on the basis of the "right to left"rule;

10 shows an intersection with traffic lights, the red light illuminating, and with a green arrow-on-black-ground sign for the direction of travel of the driver's vehicle; and.

11 shows an intersection at which the priority road makes a right turn.

According to 1 includes a driver intention prediction device 10 a crossing situation identification unit 12 ,

On its input page, which is shown on its left side, the intersection situation Identifier unit 12 with input devices 14 . 16 and 18 connected. As a non-limiting example, the input device may 14 a vehicle position detecting device, e.g. A GPS device, may be the input device 16 a map database device and may be the input device 18 a pattern recognition device which evaluates images received from a camera (not shown) provided in the vehicle V of the driver (see FIG 3 to 11 ).

The intersection situation information may include information about the stationary situation of the intersection as such.

As mentioned above, the term "stationary situation of the intersection as such" refers to characteristics of the intersection that do not change over time, for example, how many roads meet at the intersection (see eg. 3 ), and / or whether or not there are separate and / or combined traces for the possible directions, e.g. For example, a track for "left-straight-right" (see, eg. 6 . 8th and 9 ) or two tracks, namely a first track for "left-straight" and a second track for "right" (see, for example, 10 ) or a first track for "left" and a second track for "straight ahead-right" (see eg. 3 ) or three tracks, namely a first track for "left", a second track for "straight ahead" and a third track for "right" (see eg. 4 and 5 ), and / or whether or not traffic lights are present at the intersection (see eg. 3 to 6 and 10 ), and / or whether or not there are separate traffic lights for separate lanes (see eg 3 and 4 ), and / or whether or not street signs are present at the intersection (see eg 7 to 9 and 11 ), and / or what type of road signs are provided at the intersection, and similar information.

As such, the intersection situation information about the stationary situation of the intersection may be provided by the vehicle position detection device 14 and the map database device 16 to be obtained. However, you can also use the pattern recognition device 18 to be obtained.

Further, the intersection situation information may include information about the transient situation of the intersection as such.

As mentioned above, the term "transient situation of the intersection as such" refers to characteristics of the intersection, which may change over time, regardless of the traffic situation at the intersection, e.g. Whether or not the traffic lights are in operation and / or what the current state of the traffic lights is (red-yellow-green, see eg. 5 and 10 ), and similar information.

As such, the intersection situation information about the transient situation of the intersection may be provided by the pattern recognition device 18 to be provided.

Further, the intersection situation information may include information about the vehicle V of the driver approaching the intersection.

As mentioned above, the behavior of the driver may depend on the type of lane on which the driver's vehicle V approaches the intersection (see, for example, FIG. 3 to 5 and 10 ).

As mentioned above, the information about the vehicle V of the driver approaching the intersection may be obtained by, for example, the pattern recognition device 18 to be provided.

The intersection situation identification result is sent to a first selection unit 20 and a second selection unit 22 forwarded to the output side of the intersection situation identification unit 12 are connected, which are shown on the right side thereof.

Based on the intersection situation identification result obtained from the intersection situation identification unit 12 is obtained, determines the first selection unit 20 which of the prediction units 24 . 26 and 28 is most appropriate to provide a reliable prediction of driver intention at the intersection. The prediction units 24 . 26 and 28 may be formed by any known type of classification device, e.g. For example, a classifying device including a Bayesian network, a fuzzy logic-based classification device, a support vector machine-based classification device, an artificial neural network-based classification device to list only a few types of matching classification devices. It should also be noted that the driver intention prediction device 10 according to the present invention is not limited to comprising three prediction units. Instead, it may comprise less than three or more than three prediction units.

On its input page, which is shown on its left side, is the second selection unit 22 with a plurality of indicator providing units 30 . 32 and 34 which in turn are connected at their respective input sides, which are shown at their respective left sides, with input signal Supply units 36 and 38 are connected. It should be noted again that the driver intention prediction device 10 according to the present invention is not limited to comprising three indicator providing units. Instead, it may comprise fewer than three or more than three indicator providing units. Moreover, it is not limited to including two input signal providing units. Instead, it may include more than two input signal providing units.

The input signal providing unit 36 is with two indicator providing units 30 and 32 to illustrate, by way of non-limiting example, that one and the same input signal may be evaluated according to different evaluation methods to provide two or more different driver intention indicators. For example, the vehicle speed could be evaluated to provide a current value of the vehicle speed and / or a current value of the acceleration / deceleration of the vehicle.

However, it is also conceivable to evaluate a specific input signal only in a specific way to provide a specific driver intention indicator. For example, the indicator providing unit 34 receive an input signal that provides information about the head swivel angle of the driver's head, accumulates that sensor signal for a predetermined period of time or for a predetermined distance of the vehicle, and provides the average head swivel angle as one of the driver intentional indicators. The head swing angle information may be from, for example, a pattern recognition unit 38 which evaluates images of the driver taken by an on-board camera of the driver's vehicle using pattern recognition methods. Similarly, an average driver's line of vision may be provided as one of the driver intentional indicators.

Further, it should be noted that at least one indicator providing unit receives the sensor signal as from the sensor unit, i. H. without any calculation, than can provide the driver intent indicator. For example, the current gear as obtained from a gear shift sensor or the turn signal state as obtained from a turn signal sensor may be provided as one of the driver intention indicators.

From the above, it will be understood that an input signal providing unit may comprise either a sensor unit, e.g. A gear shift sensor, a turn signal sensor, a vehicle speed sensor or the like, or an evaluation unit that evaluates the output signals provided by at least one sensor unit, e.g. A pattern recognition unit that evaluates images of the driver taken by an on-board camera to provide a head swing angle.

Another type of input signal may include information about other moving objects O (see, for example, FIG. 9 and 10 ), since the behavior of the driver will depend on whether his vehicle V (see eg. 9 ) or one of the other moving objects O (see eg 10 ) has the right of way.

Based on the intersection situation identification result obtained from the intersection situation identification unit 12 is received, determines the second selection unit 22 which is indicated by the Indicator Provision Units 30 . 32 and 34 provided indicators to which of the prediction units 24 . 26 and 28 should be forwarded. This means that, depending on the crossing situation identification result, one and the same prediction unit can receive different sets of indicators.

Finally, the prediction unit returns by the first selection unit 20 is selected, the driver intention prediction result to at least one driving assistance device. As a non-limiting example, in 1 only a driver assistance device 40 shown. A first type of driver assistance device may be an active driver assistance device that actively assists the driver, for example, by activating the turn signal if the driver has failed to activate it, and / or by activating the brakes, if that There is a risk that the driver might disregard the right of way of another vehicle, whereas a second type of driver assistance may be a passive driving assistance device that passively assists the driver, for example by alerting him to a potentially dangerous situation, e.g. B. by optical and / or acoustic signals.

For completeness, it should be mentioned that the driver intention prediction device 10 according to the present invention, a triggering device 42 which is adapted to start the operation of the intersection situation identification unit 12 and therefore the entire driver intention prediction device 10 if a predetermined condition has been met. For example, the operation may be triggered if the distance to an intersection approaching the driver's vehicle is less than a predetermined distance, e.g. B. less than 20 m. Of the Distance to the intersection may be determined, for example, based on data obtained from a vehicle position detecting device and / or a map database and / or by evaluating images taken by a camera located in the driver's vehicle and / or the intersection itself if the intersection is equipped with intelligent intersection technology.

2 shows a second embodiment of a driver intention prediction apparatus according to the present invention, which is only slightly different from the driver intention prediction apparatus 10 , what a 1 shown differs. Therefore, components of the second embodiment which correspond to components of the first embodiment will be denoted by the same reference numerals as in FIG 1 but increases by 100. In addition, the driver intention prediction device becomes 110 to 2 only insofar as they differ from the driver intention prediction device 10 to 1 different. For the rest of the description is hereby explicitly to the description 1 Referenced.

The driver intention prediction device 110 , what a 2 is different from the driver intention prediction device 10 , what a 1 is shown only in that they are not a second selection unit, corresponding to the second selection unit 22 to 1 , includes. Instead, the indicator deployment units are 130 . 132 and 134 with the corresponding prediction units 124 . 126 and 128 permanently connected.

That means that each of the prediction units 124 . 126 and 128 always receives the same set of indicators provided by the indicator providing units 130 . 132 and 134 based on the input signals provided by the input signal providing units 136 and 138 be received.

With regard to triggering the operation of the intersection situation identification unit 112 through the trigger unit 142 , with reference to the identification of the crossing situation by the crossing situation identification unit 112 based on the input signals provided by the input devices 114 . 116 and 118 with respect to the selection of the most suitable prediction unit by the selection unit 120 , and with respect to the output of the prediction result to the driving assistance device 140 Please refer to the description of the embodiment 1 Referenced.

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

• JP 2010-097335A [0002]

Claims (15)

Contraption ( 10 ), which is adapted to predict a driver intention at an intersection, comprising: an intersection situation identification unit ( 20 ) which is • arranged to receive intersection-situation information, • is arranged to identify the intersection situation on the basis of the received intersection-situation information, and • is arranged to perform a intersection-situation identification • provide at least one indicator providing unit ( 30 . 32 . 34 ) configured to receive at least one sensor signal, and configured to provide a driver intent indicator based on the at least one received sensor signal, at least two prediction units, 24 . 26 . 28 each configured to provide a driver intention prediction output based on at least one driver intent indicator provided by the at least one indicator providing unit, and a selection unit ( 20 ), which is adapted to one of the prediction units ( 24 . 26 . 28 ) on the basis of the intersection-situation identification output to determine the driver-intention prediction output of the selected prediction unit as the driver-intention prediction result of the device ( 10 ). Device according to claim 1, characterized in that the intersection situation information comprises information about the stationary situation of the intersection as such. Apparatus according to claim 1 or 2, characterized in that the intersection situation information comprises information about the temporary situation of the intersection as such. Device according to one of claims 1 to 3, characterized in that the intersection situation information contains information about the driver's vehicle approaching the intersection. Device according to one of claims 1 to 4, characterized in that the intersection situation information comprises information about other moving objects. Device according to one of claims 1 to 5, characterized in that it comprises at least one of the following interfaces: • an interface for obtaining intersection situation information from a vehicle position detection device ( 14 ) and / or a map database ( 16 ), An interface for obtaining intersection-situation information from a pattern recognition device (FIG. 18 ), which receives images from a camera, • an interface for obtaining intersection situation information from an intersection equipped with an intelligent intersection technology system. Device according to one of claims 1 to 6, characterized in that the at least one indicator-providing unit ( 30 . 32 . 34 ) calculates the driver intent indicator based on the at least one received sensor signal. Device according to one of claims 1 to 7, characterized in that at least one indicator-providing unit ( 30 . 32 . 34 ) provides the sensor signal as received as the driver intent indicator. Device according to one of claims 1 to 8, characterized in that it comprises a second selection unit ( 22 ), which is adapted to the intersection situation identification output from the intersection situation identification unit ( 12 ) and the driver intention indicators from the at least two indicator providing units ( 30 . 32 . 34 ), and is arranged, based on the intersection situation identification output, at least one driver intention indicator to the prediction unit ( 24 . 26 . 28 ) provided by the first selection unit ( 20 ) is selected. Device according to one of claims 1 to 8, characterized in that each prediction unit ( 124 . 126 . 128 ) with at least one associated indicator providing unit ( 130 . 132 . 134 ) is permanently connected. Device according to one of claims 1 to 10, characterized in that at least one prediction unit ( 24 . 26 . 28 ) comprises a classification device. Device according to one of claims 1 to 11, characterized in that it is adapted to the driver intention prediction result to at least one driving assistance device ( 40 ). Device according to one of claims 1 to 12, characterized in that it comprises a triggering device ( 42 ), which in addition is set up to start the operation of the driver intention prediction device ( 10 ), if a predetermined condition has been met. Vehicle comprising a device according to one of claims 1 to 13. A method for predicting driver intent at an intersection, comprising the steps of: receiving intersection situation information, identifying the intersection situation based on the received intersection situation information, receiving at least one sensor signal, providing at least one driver intent indicator based on the at least one received sensor signal, • selecting one of a plurality of prediction methods based on the intersection situation identification output, • predicting the driver intention based on the identified intersection Situation and the at least one provided driver intent indicator using the selected prediction method.

Images