Method for operating a direction indicator device of a motor vehicle, control device, server device, and motor vehicle

Abstract

The invention relates to a method for operating a direction indicator device (16) of a motor vehicle (10), wherein a control device (12): receives (S1) traffic situation data from a motor vehicle system of the motor vehicle (10), which data describe a traffic situation in which the motor vehicle (10) is located; receives (S2) driver control data, which describe operating of the motor vehicle (10) by its driver; predicts (S3), on the basis of the received driver control data, a driver intention with respect to a driving manoeuvre planned by the driver in the described traffic situation; and determines (S8), on the basis of the traffic situation described by the received traffic situation data and on the basis of the predicted driver intention, a target control action of the direction indicator (24, 26, 28) of the direction indicator device (16), which target control action describes a control action associated with the described traffic situation of one or more direction indicators (24, 26, 28) of the direction indicator device (16). The control device (12) provides (S9) a control signal which describes the determined target control action, and transmits the provided control signal to the direction indicator device (16, S10).

Description

[0001] Description

[0002] Method for operating a direction indicator device of a motor vehicle, control device, server device, and motor vehicle

[0003] The invention relates to a method for operating a direction indicator device. A direction indicator device is understood to be a device, a device component, or a device group that is configured to use lighting technology to warn road users of a change in direction of travel, a lane change, or a hazardous situation. A direction indicator device comprises at least one, preferably several, direction indicators or turn signals for activating and deactivating a flashing light and a warning light.

[0004] The direction indicator is therefore a component of a vehicle's lighting system. However, turn signals or hazard warning lights are often used incorrectly, too late, or not at all. Typical situations include turning without signaling, changing lanes without signaling, stopping or starting without signaling, indicating an intention to overtake while the vehicle is already in the overtaking lane (i.e., tailgating), and falsely using the turn signal on a turning priority road and / or in a roundabout.Other typical situations in which a direction indicator is misused include forgetting to turn off the indicator, for example, after changing lanes, aborting a lane change, or changing direction; incorrectly using the hazard warning lights; flashing when stopped at the side of the road or in a double row without posing a danger to traffic; incorrectly using the hazard warning lights to indicate dangerous situations, for example, at the end of a traffic jam or in the event of an accident on the motorway; and / or automatically switching them on and off. Incorrect use of the direction indicator can lead to ambiguous and therefore misleading traffic situations and thus to misjudgments by other road users.

[0005] WO 2010 / 045908 A1 describes a method for automatic turn signaling, comprising the steps of detecting driver activity and / or vehicle position using a lane-keeping assistance system and / or a lane-keeping assistant, and activating or deactivating a turn signal depending on the detected driver activity or vehicle position. US 2019 / 0077308 A1 describes systems and methods for activating turn signal indicators of a motor vehicle.

[0006] An object underlying the invention is to improve the quality of control of the direction indicator device by a control device of the motor vehicle.

[0007] The stated object is achieved by the method and devices according to the invention according to the independent claims. Advantageous further developments are provided by the subordinate claims.

[0008] The invention is based on the idea of ​​predicting driver intent based on the way the driver operates the motor vehicle, rather than observing the driver, i.e., instead of determining driver behavior by observing the driver, and automatically controlling the turn signal device based on the predicted driver intent. Preferably, data can be used using vehicle-to-X communication as an input and / or output channel. Ideally, a generic, pre-trained driver intent model can be used that can provide experience data for the largest possible use case, so that this experience data can be made available to other motor vehicles. This allows any driver to benefit from the experiences of an individual driver, and a corresponding artificial intelligence can learn from this.The training data and / or experience data can preferably be made available in a cloud system. Unlike the prior art, where the driver is observed or where control is based on traffic regulations and the proximity of lane markings, the system of the invention can "get to know" a driver based on their driving behavior.

[0009] This can advantageously enable automatic adjustment of the direction indicator control, which is particularly advantageous in countries with different legal regulations. For example, the turn signal behavior in roundabouts may be regulated differently in different countries.

[0010] The method according to the invention for operating a direction indicator device of a motor vehicle is carried out by a control device. A control device is understood to be a device, a device component, or a device group that is configured to receive signals, evaluate them, and provide control signals, for example, to generate them, as well as to transmit the provided control signals to other components. For this purpose, the control device can, for example, have a corresponding receiving unit, a data processing unit, and a transmitting unit.

[0011] The control device receives traffic situation data from a motor vehicle system of the motor vehicle, wherein the received traffic situation data describe a traffic situation in which the motor vehicle is located. The traffic situation data can preferably be received via vehicle-to-X communication, for example, via vehicle-to-vehicle communication, or from a data server external to the vehicle. The received traffic situation data can also be described as driving situation data. The traffic situation data can, for example, describe a traffic volume or a traffic jam.

[0012] The control device also receives driver control data, whereby the received driver control data describes the operation of the motor vehicle by its driver. For example, the driver control data can describe a steering sequence with corresponding steering angles and steering inputs, as well as technical data on the driver's acceleration and braking behavior. The control device can preferably receive the driver control data within a predefined test period. By including the driver control data, there is no need to monitor the driver, for example, using cameras. The driver control data is also much more precise with regard to a possible driver intention. A possible driver intention can, in particular, be a driving maneuver such as a lane change or a turning maneuver.

[0013] The control unit predicts the driver's intention based on the received driver control data regarding a driving maneuver planned by the driver in the described traffic situation. This provides more precise information about the driver's intention than driver observation.

[0014] The control device determines a target control of the direction indicator device based on the traffic situation described by the received traffic situation data and the predicted driver intention. The determined target control describes a control of one or more direction indicators of the direction indicator device assigned to the described traffic situation. In other words, the determined target control describes a flashing, a deactivation of a turn signal, or an activation and / or deactivation of a hazard warning light that is appropriate or very likely appropriate given the predicted driver intention. Preferably, the control device determines the target control based only on the traffic situation data and the predicted driver intention. The target control is therefore much more precise than controlling the direction of the direction indicator based on driver observation.

[0015] The control device provides a control signal, for example, generates it, whereby the provided control signal describes the determined desired control. The control device then transmits the provided control signal to the direction indicator device.

[0016] The advantages mentioned above arise.

[0017] Preferably, the control device can provide the traffic situation and the driving control data and transmit them to a deep learning engine. The combined data can optionally be included as a training dataset for training the deep learning engine.

[0018] A deep learning engine ("deep learning device") is a device, device component, or program that can apply so-called deep learning (so-called in-depth learning, machine learning) to a wide variety of data. In other words, a deep learning engine is a highly sophisticated device for conducting in-depth learning, i.e., an implementation of artificial intelligence. In other words, both artificial intelligence as machine learning and in-depth learning can be implemented using the deep learning engine. The deep learning engine can, for example, be designed and / or configured as a deep artificial neural network.In other words, the deep learning engine can be configured to use a machine learning method to evaluate a large number of empirical values ​​and / or training data, which can also be referred to as a training dataset, or a dataset according to a predetermined algorithm and based on the large number of empirical values ​​already stored, for example, using logic contained therein, such as a correlation. This can also create further logical connections in the deep learning engine.

[0019] Empirical values ​​or training data can, for example, be statistically summarized for a multitude of different driver intentions associated with traffic situation data and driver control data. For this training step, the control device can operate the deep learning engine. The control device can operate the deep learning engine to process the provided traffic situation data and driver control data and thereby determine a driver intention forecast. This driver intention forecast includes or describes a probability with which the driver intends a driving maneuver. The control device can then operate the deep learning engine to derive a driving maneuver as the driver intention based on the driver intention forecast, which the driver intends with a predetermined minimum probability.

[0020] This embodiment of the method according to the invention significantly increases the variety of possible driving maneuvers, thus predicting the driver's intentions even more accurately. The aforementioned advantages thus increase synergistically.

[0021] In a further embodiment of the method according to the invention, the control device checks whether the received driving control data describe an actuation of the direction indicator device in accordance with the determined target control. In other words, in this embodiment, the control device checks whether the driver is correctly activating the turn signal or hazard warning lights, or whether the actuation is false positive or false negative. The control device can then only provide the control signal if the received driving control data describe an actuation of the direction indicator device that deviates from the determined target control. In other words, the control device only intervenes if the driver is using the direction indicator device incorrectly.In addition to a particularly high learning effect for the driver, who can be corrected and learn from incorrect use of the direction indicators, this can also be used as a starting point for assessing the driver's driving skills.

[0022] To effectively counteract the problem described above, the turn signal or turn signal and hazard warning lights can be switched on and / or off according to the following variants. These variants can optionally be combined with one another as desired.

[0023] In one embodiment of the method according to the invention, the predicted driver intention can describe the actuation of one of the direction indicators. The control device can then distinguish between a comfort flashing and a continuous flashing when manual actuation of the direction indicator lever is detected. A comfort flashing is understood to be a flashing sequence in which the direction indicator lever is only lightly actuated, i.e., for example, without the lever engaging, and / or in which the driver manually controls only a few repetitions of the flashing, for example, only 3-5 flashes. Typically, the lever is only lightly tapped for the comfort flashing. Optionally, the control device can, for example, detect a lane change due to a slight steering angle and derive the comfort flashing from this.

[0024] In this embodiment, the control device specifies a flashing duration (and thus the number of flashing repetitions) depending on whether the predicted driver intention describes a lane change, driving off with the motor vehicle, merging the motor vehicle into a lane, or turning. Additionally or alternatively, the control device can deactivate the activated direction indicator after the driving maneuver has ended. The control device can then control a comfort flashing, for example during a lane change, if it predicts a lane change as the driver intention, or a longer flashing duration compared to the comfort flashing if the predicted driver intention describes a turn. Optionally, a distinction can be made between before a turn and the planned turn. In another example, the control device can specify a long flashing duration despite a comfort flashing initiated by the driver.

[0025] To enable a route-based direction indicator, in another variant, the control device can determine a motor vehicle's travel route specified by a navigation device. A navigation device is understood to be a device, device component, or device group configured to create travel routes and provide navigation guidance. The navigation device can preferably be configured as a navigation device or navigation system of the motor vehicle.

[0026] The control device can then determine an upcoming change of direction based on location data of the motor vehicle, which describes the geographical position of the motor vehicle, and / or route guidance data, which describes the course of the travel route. Based on the determined change of direction, the control signal can be used to activate the turn signal. Route guidance data can be, for example, map data and / or data collected using traffic sign recognition.

[0027] Additionally or alternatively, the control device may omit, block, or revoke the provision and / or transmission of the control signal for activating the turn signal if the traffic situation does not allow the route to be safely followed, or if multiple driving maneuvers are to be performed before a turn. The route cannot be safely followed if, for example, the predicted driver intention describes a lane change into the wrong lane, if a turning lane is blocked, in cases of heavy traffic, and in a situation where multiple vehicles are traveling in parallel.

[0028] Additionally or alternatively, the control device can receive environmental observation data, preferably via vehicle-to-X communication or via its own sensors, for example via radar or LIDAR data. The environmental observation data describes the environment of the motor vehicle in the traffic situation. If a current traffic situation and / or several pending driving maneuvers must be performed before a turn, the control device can generate an indication signal and transmit it to an output device of the motor vehicle, for example to a screen and / or a loudspeaker. The generated indication signal, which can be output, for example, via a screen, can describe an indication of the several pending driving maneuvers.

[0029] Additionally or alternatively, the control device can use the received driver control data to determine or predict the completion and / or cancellation of a currently initiated driving maneuver, and control the deactivation of an activated turn indicator. For example, a currently initiated driving maneuver can be completed if the turn was successful, while the driving maneuver can be canceled if the driver does not turn after all.

[0030] Such a fusion with navigation data can optimize the control of the turn signal, for example, at construction sites where the navigation system does not yet have information that a lane has been closed due to construction. Preferably, camera data monitoring the vehicle's surroundings can also be included, allowing the control system to determine the construction site or the closed lane, for example, through image analysis.

[0031] In a further variant, the control device can support route-based control of the direction indicators based on driving intent recognition. For this purpose, the control device can receive driver observation data from a sensor device of the motor vehicle, i.e., from a device, a device group, or a device component for detecting environmental characteristics. The sensor device can preferably comprise a camera for monitoring the interior. In this variant, the control device can derive driver behavior while driving based on the received driver observation data and additionally determine the desired control of the direction indicator of the direction indicator device based on the derived driving behavior.

[0032] An example of a route-based turn signal supported by driver intent recognition based on driver observation is when the driver intends to discontinue following a route. The turn signal cannot be activated in this case or can be deactivated before the maneuver is completed. Here, too, the turn signal can optionally be deactivated automatically after the maneuver is completed.

[0033] If, for example, the driver wants to take a break and the driver's intention is, for example, to drive to a rest area, the direction indicator can be set much more precisely if he or she merges into or out of the lane late.

[0034] In one variant relating to the motor vehicle's hazard warning flashing, the control device can determine a target control based on the traffic situation that provides for the activation of the hazard warning flashing, and the provided control signal can describe the activation of the hazard warning flashing. This also optimizes the use of the direction indicator device with regard to hazard warning flashing.

[0035] Examples of this include automatic hazard warning flashers or optional deactivation of the hazard warning lights. For this purpose, data from the traffic surrounding the vehicle, as well as optional driver observation, can be used to detect, for example, the approaching of the end of a traffic jam or the vehicle itself as a traffic obstruction.

[0036] In particular, traffic density can be taken into account, which can be determined using a camera, for example, and / or information from mobile online services and / or other infrastructures. This could include information about whether there is an accident-damaged vehicle in the immediate vicinity whose driver has deployed a warning triangle.

[0037] The invention also includes the control device. The control device can have a data processing apparatus or a processor device that is configured to carry out an embodiment of the method according to the invention. For this purpose, the processor device can have at least one microprocessor and / or at least one microcontroller and / or at least one FPGA (Field Programmable Gate Array) and / or at least one DSP (Digital Signal Processor). Furthermore, the processor device can have program code that is configured to carry out the embodiment of the method according to the invention when executed by the processor device. The program code can be stored in a data memory of the processor device. The control device can preferably be designed as a control device, control chip, computer, control unit, or application program (“app”).

[0038] Preferably, the control device may comprise the deep learning engine. The deep learning engine may preferably be configured to perform the above-mentioned steps of a deep learning engine.

[0039] The above-mentioned object is achieved by a server device for operation on the Internet, for example, a data server, a backend, and / or a data cloud, wherein the server device comprises an embodiment of the storage medium according to the invention and / or an embodiment of the control device according to the invention. The server device is, in particular, a server device external to the motor vehicle.

[0040] The invention also includes a motor vehicle having a direction indicator device and an embodiment of the control device according to the invention. The motor vehicle is preferably designed as a motor vehicle, in particular as a passenger car or truck, or as a passenger bus or motorcycle.

[0041] The invention also relates to a computer or a computer program product comprising instructions which, when the program is executed by a computer, cause the computer to carry out an embodiment of the method according to the invention.

[0042] The invention also relates to a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out an embodiment of the method according to the invention.

[0043] The invention also includes further developments of the control device according to the invention, the server device according to the invention, and the motor vehicle according to the invention, which have features as already described in connection with the further developments of the method according to the invention. For this reason, the corresponding further developments of the control device according to the invention, the server device according to the invention, and the motor vehicle according to the invention are not described again here.

[0044] The invention also includes combinations of the features of the described embodiments.

[0045] Exemplary embodiments of the invention are described below. Shown are:

[0046] Fig. 1 is a schematic representation of a first embodiment of the method and devices according to the invention; and

[0047] Fig. 2 is a schematic representation of a further embodiment of the method and devices according to the invention.

[0048] The exemplary embodiments explained below are preferred exemplary embodiments of the invention. In the exemplary embodiments, the described components each represent individual, independently considered features of the invention, which also further develop the invention independently of one another and are thus also to be considered as components of the invention, either individually or in a combination other than that shown. Furthermore, the described exemplary embodiments can also be supplemented by further features of the invention already described.

[0049] In the figures, functionally identical elements are provided with the same reference numerals.

[0050] Fig. 1 illustrates the principle of the method and device according to the invention using a first exemplary embodiment. In other words, Fig. 1 shows a system overview. Fig. 1 shows a motor vehicle 10, for example, a passenger car, which may include the control device 12. The control device 12 may be configured, for example, as a control unit or control chip and may have a communication interface 14, preferably a communication interface 14 for motor vehicle-to-X communication. A mobile radio connection and / or an internet connection may preferably be used for data communication.Optionally, the motor vehicle 10 can have an antenna 15, for example, for determining the positioning of the motor vehicle, i.e., for example, for receiving location data of the motor vehicle, for example, a GPS receiver; and / or, for example, a DAB+ antenna, a radio antenna, and / or an antenna 15 for mobile online services. The antenna 15 can then preferably be part of the communication interface 14.

[0051] Within the motor vehicle 10, in particular for communication with a direction indicator device 16, the communication can take place via a wired data communication connection 18, for example a data bus of the motor vehicle 10, or also via a common wireless data communication connection 18, for example a WiFi connection or Bluetooth LE connection.

[0052] Optionally, the control device 12 may comprise a processor device 20 and / or a data memory 22.

[0053] In the example of Fig. 1, the direction indicator device 16 can preferably comprise two direction indicators 24 at the front of the motor vehicle, two side direction indicators 26, and two direction indicators 28 at the rear of the motor vehicle. For reasons of clarity, the data communication connections 18 of the direction indicators 24, 26, 28 to the direction indicator device 16 of Fig. 1 are not shown.

[0054] Fig. 1 shows an optional deep learning engine 30 as a component of the control device 12. Alternatively to the illustration in the figure, the deep learning engine 30 can also be, for example, a vehicle-external deep learning engine 30, which can, for example, be part of a vehicle-external data server (not shown in the figure). Accordingly, the data transmission between the control device 12 and a vehicle-external deep learning engine 30 can take place via, for example, a mobile phone connection or internet connection.

[0055] A sensor device 32 of the motor vehicle 10 can ideally have a front sensor system 34 with, for example, a radar and / or a LIDAR and / or an ultrasound-based sensor. Additionally or alternatively, the sensor device 32 can include a front camera 36 and, optionally, a driver observation sensor system (not shown in Fig. 1). Such an optional driver observation sensor system can be located inside the motor vehicle 10 (not visible in Fig. 1). Advantageously, the sensor device 32 can additionally include a rear sensor system 38, which can preferably have a radar and / or a LIDAR and / or a camera and / or an ultrasound-based sensor system.

[0056] In the example of Fig. 1, the control device 12 receives traffic situation data in S1, for example via the communication interface 14, which may describe, for example, a traffic jam or a busy multi-lane road. Furthermore, the control device 12 receives driver control data in S2, which may describe, for example, the driver's steering behavior, as well as longitudinal acceleration. Corresponding steering wheel sensors and acceleration sensors are known to those skilled in the art. If, for example, the motor vehicle 10 drives close to another motor vehicle traveling in front (not shown in Fig. 1) and the driver steers significantly to the left, the control device 12 can predict, for example, an overtaking maneuver as the driver's intention (S3).

[0057] Optionally, the control device 12 can transmit the traffic situation data and the driver control data to the deep learning engine 30 in S4 (S4) and then operate it (S5) to determine a driver intention forecast (S6) and to derive the overtaking maneuver based on the driver intention forecast (S7).

[0058] For the desired control of the direction indicators 24, 26, 28, the control device 12 can determine the activation of the left front direction indicator 24 (S8). The control device 12 can generate a corresponding control signal for this purpose (S9) and transmit this to the direction indicator device 16 (S10).

[0059] Optionally, it can be provided that the control device 12 checks beforehand (S11) whether the driver activates the left front turn signal on his own initiative. Since this cannot be the case, for example, in the example shown in Fig. 1, the control signal is provided by the control device 12 (S9) and transmitted to the direction indicator device 16 (S10).

[0060] If, for example, the driver has only lightly tapped the steering column switch to activate the left turn indicator 24, thereby initiating a convenience turn signal, the control device 12 can specify the duration of the turn signal based on the predicted lane change (S12), so that the left front turn indicator 24 flashes not just three to five times, but, for example, ten times, or until the motor vehicle 10 is in the left lane. Ideally, the control device 12 can then deactivate the turn indicator 24 again (S13). Using location data, which the control device 12 can receive, for example, via the communication interface 14 from a satellite or a data server external to the vehicle (not shown in Fig. 1), and based on the course of a travel route, the control device 12 can determine in S14, for example, that the motor vehicle 10 must shortly change back to the right lane and turn right.The right front turn signal 24 can then be controlled accordingly (S15). If the control device 12 receives environmental observation data (S16), these can describe, for example, a current traffic situation, such as a traffic jam, and the control device 12 can accordingly generate an indication signal indicating the traffic jam (S17) and transmit it to an output device of the motor vehicle 10, such as a screen of an instrument cluster (S18). Optionally, the traffic environment can be evaluated (S21).

[0061] If the sensor device 32 additionally receives driver observation data (S19), the driver's behavior during the journey can be derived (S20) and taken into account when determining S8 the desired control of the direction indicators 24, 26, 28. Such driver observation data can, for example, describe the driver's facial expressions or how the driver generally behaves in the motor vehicle. An optional driver observation sensor system can preferably comprise one or more interior cameras.

[0062] Fig. 2 shows an exemplary functional diagram for a further embodiment of the method according to the invention. Fig. 2 shows a server device 40 external to the motor vehicle, which can be configured, for example, as a backend or data cloud. The control device 12, which can be configured, for example, as an "Electronic Control Unit" ("ECU"), i.e., as a microcontroller or control unit, can include the deep learning engine 30, which can use artificial intelligence to evaluate the traffic accident, the driving intention, and optionally also the navigation. Alternatively, the server device 40 can have the deep learning engine 30, which is then in close communication with the control device 12.

[0063] The control device 12 can preferably take over the sensor and / or data fusion S22.

[0064] In addition to the method already described above, the data sets can also be summarized in a vehicle data set 42, which can preferably be used for user-centered learning S23. For user-centered learning S23 using the vehicle data set 42, a user-specific basic model can be provided that can specify how the deep learning engine 30 should react. The basic model can describe the driver's user preferences, learned route data for which the driver 56 correctly set the turn signal 24, 26, 28, and / or legal requirements. The deep learning engine 30 can optionally also learn additional information, for example, if the driver 56 corrects a control of the deep learning engine 30.

[0065] Input data and input signals can include, for example, data 44 relating to route planning, data from the longitudinal column switch and / or hazard warning button. Location data 46, which can be received via a location signal, for example a GPS signal, can preferably also be included. Further input data and input signals can be: sensor signals 48 from, for example, the front sensor system 34, in particular data from radar, camera, ultrasound, and / or LIDAR; data 50 provided by corresponding signals from online services and / or DAB+ traffic data; data 52 that can be transmitted and received via corresponding signals via vehicle-to-X; and sensor data 54 from a rear sensor system 38, in particular data from a rear radar, an ultrasound-based rear sensor system, a rear camera and / or LIDAR.

[0066] The evaluation from S21 can then be provided, for example, as a display S24 in, for example, an instrument cluster, and in the form of the control signal generated in S9 for activating and / or deactivating the direction indicator 24, 26, 28 or the direction indicators 24, 26, 28 and / or a hazard warning system (S25). The display S24 enables a driver 56 to be much better informed. Activating and / or deactivating the direction indicators 24, 26, 28 (S25) helps other road users 58 to adapt very effectively to the driving behavior of the motor vehicle 10.

[0067] Overall, the embodiments show how an autonomous direction indicator 24, 26, 28 and hazard warning lights, preferably based on artificial intelligence, is made possible.

[0068] In order to counteract the problem situations described above, the switching on and off of the direction indicators 24, 26, 28 and hazard warning lights can be implemented according to the following design variants, whereby the following design variants can optionally also be combined in different combinations:

[0069] 1) Automatic differentiation between comfort flashing (3-5 flashes) and (permanent) flashing:

[0070] - Turn signal lever is operated manually (without detent); and / or

[0071] The duration of the flashing is automatically regulated (lane change » short, 3-5 times, turning » long, continuous); and / or

[0072] For example, a distinction can be made between starting, stopping, changing lanes and merging shortly before a turn or the planned turn; and / or

[0073] The indicator is automatically deactivated after the completion of the maneuver (e.g. turning, lane change)

[0074] 2) Route-based direction indicator 24, 26, 28

[0075] - The journey follows a route planned using the navigation system; and / or

[0076] - At waypoints on this route where a change of direction must be provided and displayed, the direction indicator 24, 26, 28 is automatically activated, preferably with the aid of artificial intelligence; and / or

[0077] Information about the route, on the basis of which the necessity of the display can be decided, can be obtained from the map data and / or traffic sign recognition; and / or

[0078] - If the traffic and driving situation does not allow the route to be followed safely (for example, wrong lane, blocked turning lane, heavy traffic, parallel vehicles) or if several driving maneuvers have to be carried out before turning, the direction indicator 24, 26, 28 cannot be activated.

[0079] Information about the traffic situation or the number of necessary driving maneuvers before turning can be obtained using the AI ​​from a fusion of information from environmental observation (e.g. Car2X, radar, LIDAR).

[0080] In these cases, the upcoming maneuver (for example, following the route) could be pointed out early and possibly again; and / or

[0081] - If the maneuver is completed (e.g., turning successful) or aborted (e.g., no turn), the direction indicator 24, 26, 28 can be deactivated automatically. 3) Route-based direction indicator 24, 26, 28 supported by

[0082] Driving intent recognition

[0083] In comparison to the previously described, driver observation, preferably with the aid of artificial intelligence, can be included as an additional parameter in the control of the direction indicator 24, 26, 28; and / or

[0084] - If it is detected before or during a maneuver that the driver's driving intention has changed, for example, he or she does not intend to continue on the route, the direction indicator 24, 26, 28 cannot be activated or can be deactivated before the maneuver is completed; and / or

[0085] - Here too, the direction indicators 24, 26, 28 can be automatically deactivated after the maneuver has been completed.

[0086] 4) Direction indicators 24, 26, 28 according to the driving intention

[0087] In comparison to the above-described, the driving intention, preferably estimated using artificial intelligence based on vehicle sensors, can detect a planned driving maneuver at an early stage and the direction indicators 24, 26, 28 can be switched on; and / or

[0088] - For example, a distinction can be made between the planned turn and the planned lane change based on the driver's intention recognition; and / or

[0089] - The flashing process continues until the driving maneuver is completely completed.

[0090] 5) Automatic hazard warning lights

[0091] In situations where hazard warning flashing is required by road traffic regulations or is used as a common warning, preferably artificial intelligence will automatically activate and deactivate the hazard warning flashers; and / or

[0092] For this purpose, data from the traffic environment and the driver's own vehicle as well as driver observation can be used to detect, for example, the approaching of the end of a traffic jam or the driver's own vehicle as a traffic obstruction.

[0093] Example scenarios:

[0094] The autonomous switching of the direction indicator 24, 26, 28 can be used, for example, in the following situations:

[0095] The driver wants the direction indicators 24, 26, 28 to activate automatically, for example: The driver pulls over to the right to inform / warn other road users.

[0096] The driver pulls away from the side of the road to inform / warn the traffic behind.

[0097] - The driver parks out.

[0098] The driver pulls out to overtake.

[0099] It is important to safely pull back in during the overtaking maneuver.

[0100] The driver pulls back in after overtaking.

[0101] The driver wants to move out of an acceleration lane.

[0102] - The driver intends to turn.

[0103] - The driver intends to change lanes.

[0104] The driver intends to turn around.

[0105] The driver wants the direction indicators 24, 26, 28 to be deactivated automatically when, for example:

[0106] - the driver has finished reversing out.

[0107] - the overtaking maneuver is completed.

[0108] - the overtaking maneuver was aborted.

[0109] - the driver has returned from the acceleration lane.

[0110] - the driver has finished turning.

[0111] - the driver aborted the turn.

[0112] - the driver has changed lanes.

[0113] - the driver aborted the lane change maneuver.

[0114] - the driver's turning manoeuvre is completed.

[0115] - the driver aborted the turning manoeuvre.

[0116] The advantage is that the control of the direction indicators 24, 26, 28 can be automatically adapted to, for example, legal regulations in different countries (e.g., flashing in roundabouts); as well as only sensible and correct use of the direction indicators 24, 26, 28 and / or the hazard warning lights.

[0117] Optional system components (can also be combined):

[0118] - Navigation and route data (e.g. GPS supported) including route planner

[0119] - Multifunctional camera (preferably traffic sign, lane / roadway and object recognition)

[0120] - Additional vehicle sensors and information channels for monitoring the

[0121] Traffic environment, for example: Car-2-X, front and / or rear radar, LIDAR, ultrasound, and / or driver intent recognition and / or driver observation (for example, interior cameras, steering angle sensors, capacitive steering wheel, pedal and seat sensors, speed), and / or computing unit (i.e., the control device 12) with optional artificial intelligence, steering column switch for activating the direction indicators 24, 26, 28, hazard warning button, mandatory direction indicators 24, 26, 28 / turn signals on the vehicle. Control elements such as steering column switches and / or hazard warning buttons can be omitted in fully autonomous systems unless required by law.

[0122] The characteristics can be applied depending on the equipment variant (the number / type of sensors available).

[0123] Effect diagram (see Fig. 2)

[0124] The approach does not require driver observation (driver model) and offers a generic model with all information from the vehicle sensors, which always correctly activates and / or deactivates the direction indicators 24, 26, 28 based on, for example, the legal conditions and the traffic situation.

[0125] Preferably, a generic, pre-trained driver observation model can also be stored, which adapts to the driver at runtime. This could also be used to fine-tune the algorithm based on driver preferences within the framework of legal regulations.

[0126] List of reference symbols

[0127] Motor vehicle control device communication interface antenna

[0128] Direction indicator device Data communication link Processor device Data storage Direction indicator Direction indicator Direction indicator Deep learning engine Sensor device

[0129] Front sensors

[0130] Front camera Rear sensors Server device Vehicle data set Data

[0131] Data

[0132] Sensor data

[0133] Data

[0134] Data

[0135] Sensor data

[0136] driver

[0137] Road users 24 procedural steps

Claims

Patent claims Method for operating a direction indicator device (16) of a motor vehicle (10), wherein a control device (12): - receives traffic situation data from a motor vehicle system of the motor vehicle (10) that describes a traffic situation in which the motor vehicle (10) is located (S1), - receives driver control data describing the operation of the motor vehicle (10) by its driver (S2), - based on the received driver control data, predicts a driver intention with regard to a driving maneuver planned by the driver in the described traffic situation (S3), - based on the traffic situation described by the received traffic situation data and based on the predicted driver intention, a target control of the direction indicators (24, 26, 28) of the direction indicator device (16) is determined (S8), which describes a control of one or more direction indicators (24, 26, 28) of the direction indicator device (16) assigned to the described traffic situation, - provides a control signal (S9) that describes the determined target control, and - transmits the provided control signal to the direction indicator device (16) (S10). Method according to claim 1, wherein the control device (12): - provides the traffic situation data and the driver control data and transmits them to a deep learning engine (30) (S4), operates the deep learning engine (30, S5), in which different driver intentions associated with a multitude of traffic situation data and driver control data are statistically summarized, in order to process the provided traffic situation data and driver control data and thereby determine a driver intention prediction (S6), wherein the driver intention prediction includes a probability with which the driver intends a driving maneuver, and - the deep learning engine (30) is used to derive a driving maneuver as a driver intention (S7) based on the driver intention prediction, which the driver intends with a given minimum probability. Method according to any of the preceding claims, wherein the control device (12): - checks (S11) whether the received driver control data describes an actuation of the direction indicator device (16) in accordance with the determined target control, and - provides the control signal (S9) only if the received driver control data describes an actuation of the direction indicator device (16) that deviates from the determined target control. Method according to one of the preceding claims, wherein the predicted driver intention describes an actuation of one of the direction indicators (24, 26, 28), and wherein the control device (12) distinguishes between a comfort flash and a continuous flash when a manual actuation of the direction indicator lever is detected, and: - depending on whether the predicted driver intention describes a lane change, starting the motor vehicle (10), merging the motor vehicle (10) into a lane, or a turning maneuver: specifies a duration of the flashing (S12); and / or deactivates the activated turn signal (24, 26, 28) after completion of the driving maneuver (S13). Method according to one of the preceding claims, wherein the control unit (12) determines a route of travel of the motor vehicle (10) specified by a navigation device, and: - based on location data of the vehicle (10), which describes a geographical position of the vehicle, and / or route guidance data, which describes a course of the journey route, determines an upcoming change of direction (S14), and, based on the determined change of direction, controls the activation of the direction indicator (24, 26, 28) by means of the provided control signal (S15); and / or - refrains from providing (S9) and / or transmitting (S10) the control signal to activate the direction indicator (24, 26, 28) if the traffic situation does not allow the route to be followed safely, or if several maneuvers are to be performed before a turn, and / or - receives environmental observation data (S16) describing the environment of the motor vehicle (10) in the traffic situation; and, if a current traffic situation and / or several pending driving maneuvers are to be carried out before a turn, generates a warning signal (S17) and transmits it to an output device of the motor vehicle (10) (S18), wherein the generated warning signal describes a reference to the several pending driving maneuvers; and / or - determines or predicts the completion and / or cancellation of a currently initiated driving maneuver based on the received driver control data, and controls the deactivation of an activated turn signal indicator (24, 26, 28). Method according to one of the preceding claims, wherein the control unit (12): receives driver observation data from a sensor device (32) of the motor vehicle (10) (S19), - derives a driver behavior of the driver during the journey based on the received driver observation data (S20), and, - the target control of the direction indicators (24, 26, 28) of the direction indicator device (16) is additionally determined based on the derived driver behavior (S8). Method according to one of the preceding claims, wherein the control device (12) determines a target control (S8) based on the traffic situation, which provides for the activation of hazard warning lights, and wherein the provided control signal describes the activation of the hazard warning lights. Control device (12) configured to perform a method according to one of the preceding claims. Control device (12) according to claim 8, comprising a deep learning engine (30) configured to perform the steps relating to the deep learning engine (30). Server device (40) for operation on the Internet, comprising a control device (12) according to claim 8 or 9.Motor vehicle (10) comprising a direction indicator device (16) and a control device (12) according to claim 8 or 9.

Images