Prediction device for predicting a traffic light phase of a traffic signal system, client device and method for operating a prediction device

Abstract

The invention relates to a prediction device (12) for predicting a traffic light phase of a traffic signal system (20), wherein a map of the surroundings is stored in the prediction device (12) that describes at least one lane and a position of the traffic signal system (20). The prediction device (12) is configured to receive driving data (18) from at least one detection vehicle (16), wherein the driving data (18) of the respective detection vehicle (16) includes the vehicle position of the detection vehicle (16) at at least one detection time. The prediction device (12) is configured to determine at least one lane segment relevant to the traffic signal system (20) and, based on the driving data (18) relating to the relevant lane segment, to determine a first traffic light phase of the traffic signal system (20) at a first traffic light phase time.

Description

[0001] The invention relates to a prediction device for predicting a traffic light phase of a traffic signal system, a client device and a method for operating a prediction device.

[0002] Today's vehicles are already capable of reacting to the road ahead, i.e., the route the vehicle will soon travel, using predictive route data stored in the navigation system, and thus, for example, influencing a driving strategy in an energy-efficient manner. Furthermore, connected vehicles (V2X) also receive real-time information about their surroundings (traffic, hazards, weather, etc.). Their reactions to this information are diverse and are selected based on many factors. One basis for these decisions is the most probable path (MPP). For example, the driver is prompted to reduce speed if a dangerous curve or a temporary hazard (e.g., the end of a traffic jam or black ice) is approaching, which can only be safely navigated by reducing speed.The primary focus here is on timely intervention to manage the situation as efficiently and safely as possible and to avoid abrupt braking or dangerous situations. The same applies, for example, to a hidden speed limit, which can be approached as efficiently as possible by recognizing the situation in time. The same is true for signaling systems such as traffic lights and level crossings.

[0003] If the current or anticipated route contains non-networked traffic lights with an unknown phase, the decision regarding the reaction (e.g., reducing speed to recover energy or de-escalating the dangerous situation) rests with the driver and their foresight, perception, and experience. Without assistance from the vehicle or its sensors, the probability of an ideal reaction and thus increased efficiency is rather low. While some traffic lights are already networked with other road users via a communication module, the number of such networked traffic lights is still too small to be considered state of the art. Comprehensive networking of all traffic lights is not expected in the coming years.

[0004] US Patent 2020 066 148 A1 describes a method for the cognitive analysis and notification of traffic light phases. The method involves vehicle-to-vehicle communication of traffic light phases between surrounding vehicles to optimize traffic flow by informing drivers and / or autonomous vehicles about the phases of relevant traffic lights along a route.

[0005] DE 10 2008 015 046 A1 describes methods and devices for the predictive control and / or regulation of a hybrid drive in a motor vehicle. The method includes the selection of an operating strategy for the motor vehicle taking into account information, for example regarding traffic light phases, which is generated by communication between the motor vehicle and stationary and / or moving objects, for example by means of vehicle-to-vehicle communication or vehicle-to-infrastructure communication.

[0006] The present invention is based on the objective of providing a vehicle with a status of a traffic signal system, even if the traffic signal system is not networked.

[0007] The problem is solved by the subject matter of the independent patent claims. Advantageous further developments of the invention are described by the dependent patent claims, the following description, and the figures.

[0008] A first aspect of the invention relates to a prediction device for predicting the phase of a traffic signal system. The prediction device stores a map of the surroundings. This map describes at least one lane and at least one geographical position of a traffic signal system along that lane. In other words, the lane or lanes and one or more of the traffic signal positions are stored in the prediction device. The prediction device can, for example, be located in a central processing unit external to the vehicle to perform the prediction of the traffic signal phase for one or more ego vehicles. Alternatively, the prediction device can be located within the ego vehicle itself to perform the prediction of the traffic signal phase locally for that vehicle.

[0009] The prediction device is configured to receive driving data from the recording vehicles, whereby the driving data of each recording vehicle includes the geographical vehicle positions of the recording vehicle at the respective recording times. In other words, the recording vehicles are configured to record their vehicle positions at the respective recording times and transmit them to the prediction device.

[0010] For example, the detection vehicles can transmit the driving data to the prediction device via a mobile radio network. Alternatively, the driving data can be transmitted by detection vehicles within the vicinity of the ego vehicle to the prediction device in the ego vehicle via ad-hoc connections. The prediction device is configured to identify at least one geographically relevant road segment for the traffic signal system. This means the traffic signal system is assigned to the at least one relevant road segment. The at least one relevant road segment can, for example, be a road segment for which the traffic signal system is intended to issue traffic signals. The relevant road segment can also be a road segment that is not intended for the traffic signal system but is affected by it, e.g.,a section of roadway that crosses the section of roadway assigned to the traffic signal system.

[0011] The prediction device is designed to determine the driving data pertaining to the relevant road segment. For example, the prediction device is designed to determine the driving data encompassing the vehicle positions within the traffic signal segment at a given detection time within a specific time window. In other words, the prediction device is designed to determine the driving data related to the traffic signal phase.

[0012] The prediction device is designed to determine, based on driving data relating to the relevant road segment, the traffic signal phase at a given initial phase. This initial phase can be current or past. For example, it can be determined from the vehicle positions and the recording times of the driving data whether the vehicle being recorded is stationary or moving at the initial phase. It can also be designed to infer the traffic signal phase from the vehicle's position, for example, if the vehicle is located at an intersection beyond the traffic signal. The traffic signal phase can be determined, for example, based on a measured speed.For example, if at the first of the detection times the vehicle position of the detection vehicle is in front of the traffic light system and at a later of the detection times the vehicle position is behind the traffic light system, it can be determined that the traffic light system has a green phase at the first traffic light phase time between the two detection times.

[0013] The invention offers the advantage that the first traffic light phase can be determined from the driving data of the detection vehicles at the time of the first traffic light phase. The first traffic light phase can thus be provided to an ego vehicle even if the traffic signal system in question is not configured to provide the first traffic light phase.

[0014] The invention also includes further training.

[0015] A further development of the invention provides that the prediction device is configured to predict a second traffic light phase of the traffic signal system at a predetermined second traffic light phase time, based on the first traffic light phase of the traffic signal system at the first traffic light phase time. The second traffic light phase time can describe a future traffic light phase time that lies after the first traffic light phase time. In other words, the prediction device is configured to determine the second traffic light phase that the traffic signal system will exhibit at the second traffic light phase time from the respective first traffic light phase that was determined at the first traffic light phase time or several of the first traffic light phase times.For example, it may be provided that the second traffic light phase time is queried by a client device of the ego vehicle from the prediction device and describes an expected second traffic light phase time at which the ego vehicle will pass a certain section of the traffic light system.

[0016] A further development of the invention provides that the prediction device is configured to determine the probability of the predicted second traffic light phase. In other words, the prediction device determines the probability that the traffic signal system will exhibit the predicted second traffic light phase at the time of the second traffic light phase. This probability can be based, for example, on the quantity or accuracy of the driving data used as the basis for determining the second traffic light phase. The probability can also depend on the number of vehicles that provided the driving data. The probability can be transmitted from the prediction device to the client device. This further development has the advantage that, for example, a decision regarding the initiation of a braking maneuver can be made based on this probability.

[0017] A further development of the invention provides that the prediction device is configured to send a prediction message, comprising the first traffic light phase of the traffic signal system at the specified time and / or the second traffic light phase of the traffic signal system at the specified time, to a client device of an ego-vehicle. In other words, the traffic light phase at the first and / or second traffic light phase time is transmitted to the client device in the prediction message. The invention offers the advantage that the first and / or second traffic light phase of the traffic signal system can be provided to the client device by the prediction device even if the traffic signal system itself is not configured to do so.

[0018] A further development of the invention provides that the prediction device is configured to determine a signal timing schedule of the traffic signal system from the first traffic light phase to at least one initial traffic light phase time. In other words, the prediction device is configured to determine the signal timing schedule, which describes the relationship between the traffic light phase and time, from one or more initial traffic light phase times and the respective initial traffic light phases of the traffic signal system. This allows the traffic light phases for future traffic light phase times to be determined even without the availability of relevant driving data.

[0019] A further development of the invention provides that the prediction message includes the signal timing schedule of the traffic signal system. In other words, the signal timing schedule determined by the prediction device is provided to the client device. This offers the advantage that the client device can use the signal timing schedule to determine the traffic light phases of the traffic signal system at future times.

[0020] A further development of the invention provides that the prediction device is configured to determine the presence of a traffic signal system based on the driving data. In other words, the prediction device can deduce from the driving data that a traffic signal system is located at a given position, even if it is not stored in the map. For example, it may be observed that vehicles detecting traffic are braking at a particular position because red and green light phases are displayed there. The prediction device can then add the traffic signal system at that position to the map.

[0021] A further development of the invention provides that the driving data includes kinematic data of the respective recording vehicle at the respective recording time. In other words, the driving data describes, for example, the speed, longitudinal acceleration, and / or lateral acceleration of the respective recording vehicle. It may be provided, for example, that at each recording time, in addition to the current position of the recording vehicle, the current movement of the recording vehicle is also provided.

[0022] A further development of the invention provides that the driving data includes control data of the respective recording vehicle. In other words, the driving data describes the vehicle's control at the time of recording. The control data can, for example, describe the steering angle and / or the angle of the brake pedal and / or the angle of the accelerator pedal at the time of recording. From the control data, it can be deduced, for example, whether the recording vehicle initiates a braking process, an acceleration process, or a steering process, even if this cannot be derived from the kinematic data of the driving data at the time of recording.

[0023] A further development of the invention provides that the driving data includes a direction of travel indicated by a turn signal of the detection vehicle. In other words, the driving data describes whether a direction of travel is indicated by the turn signal of the detection vehicle at the relevant detection time. For example, the driving data can describe whether the turn signal is off, indicating a left turn, or indicating a right turn at that detection time. This has the advantage that, based on the indicated direction of travel, the detection vehicle can be assigned to a lane of the road section or to a direction of the traffic signal system.

[0024] A second aspect of the invention relates to a client device configured to receive a prediction message from a prediction device, which includes a predicted second traffic light phase of a traffic signal system at a specific time. The client device is configured to perform a predetermined reaction action depending on the second traffic light phase predicted in the prediction message. For example, the client device may be configured to issue a warning signal to the driver of the vehicle as the reaction action. It may, for example, be provided that the client device is configured to control a user interface. For example, it may be provided that the driver of the vehicle is instructed via the warning signal to brake or accelerate, depending on whether the second traffic light phase is red or green.

[0025] A further development of the invention provides that the client device is configured to send a request message to the prediction device, wherein the request message includes a request for the prediction device to provide the second traffic light phase at the second traffic light phase time. The client device can, for example, be configured to determine the second traffic light phase time at which the vehicle, based on the current movement of the ego vehicle, will reach an intersection equipped with a traffic light system. To enable predictive driving, the client device can request the second traffic light phase from the prediction device at the second traffic light phase time if the second traffic light phase cannot be provided by the traffic light system.

[0026] A further development of the invention provides that the client device is configured to adapt a determined route depending on the predicted second phase of the traffic signal. In other words, the client device can be equipped with a navigation device. The vehicle control unit can modify the determined route if, for example, following the original route, the second phase of the traffic signal is expected to be red when passing through. This has the advantage of enabling the determination of a more efficient route.

[0027] A further development of the invention provides that the client device is configured to adapt vehicle control depending on the predicted second phase of the traffic light system. For example, the client device can be configured for fully or at least partially autonomous control of the ego-vehicle. The client device can adapt the control of the ego-vehicle depending on the predicted second phase of the traffic light system and, for example, maintain the current speed of the ego-vehicle despite a current red light phase if a green light phase is predicted for the second phase.

[0028] A third aspect of the invention relates to a method for operating a prediction device for predicting a traffic light phase of a traffic signal system. The method comprises the following steps to be performed by the prediction device.

[0029] One step involves receiving driving data from the detection vehicles, where the driving data of the respective detection vehicle includes the vehicle positions of the detection vehicle at the respective detection times. One step involves determining at least one roadway segment relevant to the traffic signal system. One step involves determining the first traffic light phase of the traffic signal system at a first traffic light phase time based on the driving data along the relevant roadway segment.

[0030] For use cases or application situations that may arise during the procedure and are not explicitly described here, it may be provided that, according to the procedure, an error message and / or a request for user feedback is issued and / or a default setting and / or a predetermined initial state is set.

[0031] The invention also includes further developments of the method according to the invention, which have features already described in connection with the further developments of the prediction device and the client device according to the invention. For this reason, the corresponding further developments of the method according to the invention are not described again here.

[0032] The prediction device and the client device can include a data processing device or a processor circuit configured to perform an embodiment of the method according to the invention. For this purpose, the processor circuit can include 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). In particular, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an NPU (Neural Processing Unit) can be used as the microprocessor. Furthermore, the processor circuit can include program code configured to perform the embodiment of the method according to the invention when executed by the processor circuit. The program code can be stored in a data memory of the processor circuit. The processor circuit can, for example,based on at least one circuit board and / or at least one SoC (System on Chip).

[0033] The invention also includes combinations of the features of the described embodiments. The invention therefore also includes realizations that each exhibit a combination of the features of several of the described embodiments, provided that the embodiments have not been described as mutually exclusive.

[0034] The following are exemplary embodiments of the invention described. This is illustrated by: Fig. 1 a schematic representation of a prediction of a traffic light phase of a traffic signal system; Fig. 2. A further schematic representation of a prediction of a traffic light phase of a traffic signal system; and Fig. 3 a schematic representation of a method for operating a prediction device for predicting a traffic light phase of a traffic signal system.

[0035] The exemplary embodiments described below are preferred embodiments of the invention. In these exemplary embodiments, the described components each represent individual features of the invention, which can be considered independently of one another and each further develops the invention independently. Therefore, the disclosure is intended to include combinations of features of the embodiments other than those shown. Furthermore, the described embodiments can also be supplemented by further features of the invention already described.

[0036] In the figures, identical reference symbols denote functionally equivalent elements.

[0037] Fig. Figure 1 shows a schematic representation of a prediction of a traffic light phase of a traffic signal system.

[0038] In Fig. Figure 1 shows an ego-vehicle 10, which can move along a route towards an intersection 24 equipped with a traffic signal 20. The ego-vehicle 10 may have a client device 14, which may be configured as a navigation device. The navigation device may have determined a route for the ego-vehicle 10 and a second traffic light phase at which the ego-vehicle 10 will be located in a section 22 of the intersection 24 associated with the traffic signal 20. The client device 14 does not have access to the current first traffic light phase because the traffic signal 20 is not networked and the first phase of the traffic signal 20 is not detected by a sensor on the ego-vehicle 10.

[0039] In order to display the second traffic light phase to a driver of the Ego vehicle 10 and / or to adjust the route, the client device 14 can be configured to transmit a request message containing the second future traffic light phase time to a prediction device 12 in order to request the second traffic light phase of the traffic signal system 20 from the prediction device 12 at the second future traffic light phase time.

[0040] The prediction device 12 can be configured to predict traffic light phases of the traffic signal system 20 based on driving data 18 provided by detection vehicles 16. The prediction device 12 can be located in the ego-vehicle 10 itself or externally in a server, to which the client device 14 can be connected via a mobile internet connection. In the shown Fig. 1 The prediction device 12 is located in the Ego vehicle 10 itself.

[0041] The prediction device 12 can be configured to receive the driving data 18 from detection vehicles 16 in the vicinity of the ego vehicle 10. The driving data 18 can be transmitted by the detection vehicles 16 at specific intervals, triggered by a specific event, or upon receipt of a message by the detection vehicles 16. The driving data 18 can include detection times and current vehicle positions of the detection vehicles 16, as well as kinematic and control data. If the driving data 18 is provided by the detection vehicles 16 upon request by the prediction device 12, it can be provided that the driving data 18 only includes the detection times and vehicle positions requested in the request.The requested recording times can be limited to a current time period and the vehicle positions to sections that are relevant for the traffic signal system 20 in question, for example, those located in the vicinity of the traffic signal system 20.

[0042] It may be provided that, in the case that the prediction device 12 is set up as a server outside the ego vehicle 10, the driving data 18 is continuously transmitted to the prediction device 12 via a mobile internet connection by the recording vehicles 16 without explicit request.

[0043] The prediction device 12 can be configured to identify the sections relevant to the traffic signal system 20 and to select and evaluate the associated driving data 18. In the figure shown, one of the detection vehicles 16 is located in the relevant section of the traffic signal system 20, which may be a lane 22. The driving data 18 provided by the detection vehicle 16 indicate that the vehicle 16 is stationary at a given detection time. The prediction device 12 can thus determine that the first traffic light phase is red at the current first traffic light phase time.

[0044] The driving data 18 can also indicate for previous recording times that the recording vehicle 16 is stationary and therefore the relevant first traffic light phases are also red. From a known signal timing plan, the prediction device 12 can determine which second traffic light phase will be present at the second traffic light phase time. For example, a green second traffic light phase can be determined. The second traffic light phase can be transmitted by the prediction device 12 to the client device 14 in a prediction message 26. Based on the determined second traffic light phase, the client device 14 can maintain the determined route and display the second traffic light phase to the driver on a display device.

[0045] Fig. Figure 2 shows another schematic representation of a prediction of a traffic light phase of a traffic signal system 20.

[0046] As in Fig. 1. The ego vehicle 10 can move towards intersection 24. The section of road before intersection 24 can include three lanes 22. The traffic signal system 20 provides separate traffic light phases for the three lanes 22. The three lanes 22 can include one lane 22 for driving straight ahead, one left-turn lane, and one right-turn lane. In contrast to the one in Fig. In the situation shown in 1, it is necessary to determine the lane-specific traffic light phases.

[0047] At intersection 24, there are three of the detection vehicles 16, each positioned in one of the lanes 22. The detection vehicles 16 provide the respective driving data 18 to the prediction device 12. The driving data 18 can include the direction of travel indicated by a turn signal of the respective detection vehicle 16. Furthermore, the driving data 18 can include the orientation of the respective detection vehicle 16 as well as control data, which can, for example, specify the steering angle.

[0048] The detection vehicle 16 in the left-turn lane can indicate a left turn with its turn signal. Additionally, the steering wheel can indicate a counterclockwise steering angle. The corresponding information can be provided to the prediction device 12 in the driving data 18. From the indicated direction of travel, the steering angle, and the vehicle position, the prediction device 12 can determine that the detection vehicle 16 is in the left-turn lane or at least intends to make a left turn. The detection vehicle 16 may be stationary and not accelerating at the time of detection. The control data may indicate that the accelerator pedal of the detection vehicle 16 is not depressed and that the handbrake of the detection vehicle 16 is simultaneously activated. The prediction device 12 can therefore deduce that the detection vehicle 16 is subject to a red traffic light phase.The first traffic light phase at the first current traffic light phase time of traffic signal system 20 is therefore red.

[0049] The second of the detection vehicles 16 can be stationary in the middle lane 22 and its driving data 18 also indicates that it is stationary. The turn signal of the detection vehicle 16 can be deactivated, and the steering wheel of the detection vehicle 16 may not have any steering angle from its neutral position. From this, the prediction device 12 can deduce that the detection vehicle 16 is stationary in the middle lane 22 and that the traffic light phase for straight-ahead travel will have a first red phase at the first traffic light phase time.

[0050] The third of the detection vehicles 16 can be located in the third lane 22. The third of the detection vehicles 16 can move and accelerate. At the same time, the steering wheel of the detection vehicle 16 can be turned clockwise and the accelerator pedal of the detection vehicle 16 can be depressed. The turn signal of the detection vehicle 16 can indicate a right turn. From the provided driving data 18, the prediction device 12 can deduce that the first traffic light phase is green for the right-turn lane at the first traffic light phase time. Based on the different determined first traffic light phases for the different directions of travel, the prediction device 12 can deduce that the traffic signal system 20 displays different traffic light phases for the different directions of travel.

[0051] The signal timing schedule of traffic signal system 20 can be known to the prediction device 12. Based on the first traffic light phases of traffic signal system 20 and the signal timing schedule, the prediction device 12 can estimate which second traffic light phase will be present at the predicted second traffic light phase time. The second traffic light phase can be determined with a probability and provided to the client device 14.

[0052] The route provided by the client device 14 via a navigation function may include a right turn. The second traffic light phase may indicate that a red light will be present for the right-turn lane at the time of the second traffic light phase. The client device 14 can therefore prompt the driver, via a voice output, to gradually reduce the speed of the ego-vehicle 10.

[0053] Fig.Figure 3 shows a schematic representation of a method for operating a prediction device 12 for predicting a traffic light phase of a traffic signal system 20.

[0054] A first step S1 can include receiving driving data 18 from recording vehicles 16 by a prediction device 12, wherein the driving data 18 of the respective recording vehicle 16 includes vehicle positions of the recording vehicle 16 at the respective recording times.

[0055] A second step S2 can include the determination of at least one roadway section relevant to the traffic signal system 20 by the prediction device 12.

[0056] A third step S3 can include determining a first traffic light phase of the traffic signal system 20 at a first traffic light phase time based on the driving data 18 along the relevant roadway section by the prediction device 12.

[0057] In a fourth step S4, the prediction device 12 can predict a second traffic light phase of the traffic signal system 20 at a predetermined second traffic light phase time, based on the first traffic light phase of the traffic signal system 20 at the first traffic light phase time and a signal timing plan of the traffic signal system 20. The prediction device 12 can thereby determine a probability of the second traffic light phase.

[0058] In a fifth step S5, the prediction device 12 can send a prediction message 26, comprising the first traffic light phase of the traffic signal system 20 at the first traffic light phase time and the second traffic light phase of the traffic signal system 20 at the specified second traffic light phase time, to a client device 14 of an ego vehicle 10.

[0059] The traffic light phase of a non-networked / unintelligent traffic signal system 20 can be predicted by means of communication with other networked and intelligent road users, so-called detection vehicles 16, and aggregation of the resulting available driving data 18. With knowledge of the current traffic light phase of the traffic signal system 20, the ego vehicle 10 can react in various ways: Intelligent functions for increasing efficiency and energy recovery can react analogously to a tight curve, an obscured speed limit, and also to traffic signal systems 20 that delay or interrupt the journey to the relevant second traffic light phase.

[0060] In manual control mode, the driver of the Ego vehicle 10 can be informed via the user interface about the expected second traffic light phase of the next traffic signal system 20 at the time of the second traffic light phase.

[0061] In partially automated driving, the Ego vehicle 10 can react to the second traffic light phase of traffic light 20 at an early stage, for example before an environment detection device detects the traffic light system 20.

[0062] The navigation system can optimize the calculation of the remaining travel time or the determined route depending on the second traffic light phase.

[0063] Temporary traffic signal systems 20 can be detected early and following traffic can be informed or warned.

[0064] An estimate of how long the "red" traffic light phase typically lasts can be determined and distributed.

[0065] The second traffic light phase of unconnected traffic signal systems 20 can be predicted, which enables more energy- and time-efficient as well as safer driving.

[0066] Based on the following information, but also further information about networked vehicles in the area of ​​a traffic signal system 20, the second traffic light phase of the traffic signal system 20 can be predicted: 1. Vehicle position of the recording vehicle 16, including GPS position and any possible corrections 2. Vehicle orientation GPS of the recording vehicle 16 3. Vehicle speed of the recording vehicle 16 4. Lateral and longitudinal vehicle acceleration of the recording vehicle 16 5. Indicated direction of travel of the turn signal 6. Accelerator pedal position 7. Brake pedal position 8. Steering wheel angle / acceleration

[0067] Based on further information such as the driver's direction of gaze or the number of connected vehicles contributing to the prediction, an accuracy of the prediction can also be calculated.

[0068] In the case of multi-lane intersections 24 with specific signaling systems for each lane, the described prediction can also be carried out lane 22-selectively.

[0069] For example, navigation map data, the position and the indicated direction of travel of the turn signals of the detection vehicles 16 can provide information about the type of intersection 24 and the available lanes. The position, orientation and indicated direction of travel of the turn signal (left, right, off) of the transmitting detection vehicle 16 can provide information about which lane it is in.

[0070] Overall, the examples show how intelligent vehicle control is provided by predicting the status of a signaling system. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 2020 066 148 A1

[0004] DE 10 2008 015 046 A1

[0005]

Claims

[1] Prediction device (12) for predicting a traffic light phase of a traffic signal system (20), wherein a map of the surroundings is stored in the prediction device (12) which describes at least a roadway and a position of the traffic signal system (20), characterized by , that the prediction device (12) is configured to, - to receive driving data (18) from at least one recording vehicle (16), wherein the driving data (18) of the respective recording vehicle (16) includes a vehicle position of the recording vehicle (16) at at least one recording time; - to identify at least one section of the roadway relevant to the traffic signal system (20); and - to determine, based on the driving data (18) relating to the relevant road section, a first traffic light phase of the traffic signal system (20) at a first traffic light phase time. [2] Prediction device (12) according to claim 1, characterized by , that the prediction device (12) is configured to, - based on the first traffic light phase of the traffic signal system (20) at the one first traffic light phase time or the first traffic light phases of the traffic signal system (20) at the first traffic light phase times, to predict a second traffic light phase of the traffic signal system (20) at a given, second traffic light phase time. [3] Prediction device (12) according to claim 2, characterized by , that the prediction device (12) is configured to determine a probability of the predicted second traffic light phase at the second traffic light phase time. [4] Prediction device (12) according to claim 2 or 3, characterized by , that the prediction device (12) is configured to, - to send a prediction message (26) comprising the first traffic light phase of the traffic signal system (20) at the first traffic light phase time and / or the second traffic light phase of the traffic signal system (20) at the specified second traffic light phase time to a client device (14) of an ego vehicle (10). [5] Prediction device (12) according to any one of the preceding claims, characterized by , that the prediction device (12) is configured to determine a signal timing plan of the traffic signal system (20) from the first traffic light phases of the traffic signal system (20) at the first traffic light phase times. [6] Prediction device (12) according to claim 5 with reference to claim 4, characterized by , that the prediction message (26) includes the signal timing plan of the traffic signal system (20). [7] Prediction device (12) according to any one of the preceding claims, characterized by, that the prediction device (12) is configured to determine the presence of the traffic signal system (20) based on the driving data (18). [8] Prediction device (12) according to any one of the preceding claims, characterized by , that the driving data (18) include kinematic values ​​of the respective recording vehicle (16). [9] Prediction device (12) according to any one of the preceding claims, characterized by , that the driving data (18) include control data of the respective recording vehicle (16). [10] Prediction device (12) according to any one of the preceding claims, characterized by , that the driving data (18) include a direction of travel indicated by a direction indicator of the recording vehicle (16). [11] Client device (14), characterized by , that the client device (14) is configured to, - to receive a prediction message from a prediction device (12) which includes a predicted second traffic light phase of a traffic signal system (20) at a second traffic light phase time, and - to perform a predetermined reaction action depending on the predicted second traffic light phase. [12] Client device (14) according to claim 11, characterized by , that the client device (14) is configured to send a request message to the prediction device (12), the request message comprising a request to the prediction device (12) to provide the second traffic light phase at the second traffic light phase time. [13] Client device (14) according to claim 11 or 12, characterized by , that the client device (14) is configured to adapt a determined route depending on the predicted second traffic light phase of the traffic signal system (20). [14] Client device (14) according to any one of claims 11 to 13, characterized by , that the client device (14) is configured to adapt vehicle guidance depending on the predicted second traffic light phase of the traffic signal system (20). [15] Method for operating a prediction device (12) for predicting a traffic light phase of a traffic signal system (20), comprising the following steps to be performed by the prediction device (12): - Receiving driving data (18) from recording vehicles (16), wherein the driving data (18) of the respective recording vehicle (16) includes vehicle positions of the recording vehicle (16) at the respective recording times; - Identify at least one section of the roadway relevant to the traffic signal system (20); and - Determining a first traffic light phase of the traffic signal system (20) at a first traffic light phase time based on the driving data (18) along the relevant roadway section.

Images