Procedure for issuing notices to a vehicle user and vehicle

Abstract

Method for issuing instructions to a vehicle user, wherein the time for issuing an instruction is varied depending on the fulfillment or non-fulfillment of trigger conditions (BED1, BED2), wherein - at least two hint-specific trigger conditions (BED1, BED2) are defined to issue a specific hint; - as soon as a first triggering condition (BED1) is fulfilled, the respective triggering condition (BED1) is referred to as the first triggering condition (BED1) and the at least one further triggering condition (BED2) is referred to as the second triggering condition (BED2) and a computing unit determines a probability (P) for the fulfillment of the second triggering condition (BED2); - the computing unit compares the probability (P) with a predetermined trigger condition-specific probability threshold value (PSW); characterized in that - the computing unit, depending on the comparison, causes the output of the indication at the time (t1) when the first trigger condition (BED1) is fulfilled and suppresses it at the time (t2) when the second trigger condition (BED2) is fulfilled; or - the computing unit, depending on the comparison, suppresses the output of the indication at the time (t1) when the first trigger condition (BED1) is fulfilled and causes it at the time (t2) when the second trigger condition (BED2) is fulfilled; and wherein - the computing unit predicts a temporal progression of the probability (P), determines from the temporal progression when the probability (P) crosses the predetermined trigger condition-specific probability threshold (PSW), and suppresses the output of the indication when the first trigger condition (BED1) is fulfilled and causes it to be output when the second trigger condition (BED2) is fulfilled if a crossing duration lasting from the current moment to the crossing time is greater than a specified duration threshold and otherwise causes the output of the indication when the first trigger condition (BED1) is fulfilled.

Description

[0001] The invention relates to a method for issuing instructions to a vehicle user according to the type defined in more detail in the preamble of claim 1 and to a vehicle for carrying out the method.

[0002] To assist vehicle occupants, especially the driver, it is common practice to provide information in a vehicle. These information can be visual, acoustic, or haptic. Usually, a combined output is used, for example, simultaneous visual and auditory information.

[0003] For example, if the fuel level reaches a critical level, a corresponding warning light in the instrument cluster can illuminate, accompanied by a warning tone. If the vehicle is equipped with proximity radar, a warning tone can also be emitted if the driver is approaching a vehicle in front too quickly, and a corresponding warning message, such as a warning symbol, can be displayed on a display in the vehicle. Comprehensive information can be provided via a graphical user interface, for example, in the infotainment system.

[0004] The issuing of warnings in vehicles should be done at an optimal time so that vehicle occupants have maximum receptivity to the information and are not unduly distracted by the information in a critical traffic situation. The issuing of irrelevant warnings, especially at an inappropriate time, can also be disruptive to vehicle occupants.

[0005] To determine whether a warning needs to be issued, vehicles typically consider currently collected information, such as sensor data or status information read from a control unit.

[0006] An improved method for controlling the output of messages in vehicles is known, for example, from DE 103 40 870 A1. The method provides for adjusting the timing of the output of messages in vehicles so that an optimal output time is selected for each message. This makes the corresponding message more easily perceived by vehicle occupants and reduces distraction. The messages to be output in the vehicle are assigned a priority, and this priority is calculated using an evaluation algorithm with several criteria. The criteria are collected depending on the currently available sensor data. The document lists relevant criteria as examples of the current driving situation, traffic conditions, weather conditions, the respective road conditions, the respective driver condition, the driver type, and usage preferences.

[0007] Furthermore, DE 10 2014 203 807 A1 discloses a driver assistance system and a method for delaying or suppressing the issuance of a warning by an assistance system of a vehicle. Several triggering conditions are taken into account depending on estimated probabilities for issuing or suppressing the warning, so that the time at which the warning is issued varies.

[0008] Furthermore, DE 103 02 060 B4 discloses a method for operating a motor vehicle. The method involves estimating the probability of the motor vehicle arriving at a target area, and issuing an operating recommendation to the driver depending on whether the probability of arrival exceeds a specified threshold. Thresholds can be specified manually.

[0009] The present invention is based on the object of providing an improved method for issuing instructions to a vehicle user.

[0010] According to the invention, this object is achieved by a method for outputting information to a vehicle user having the features of claim 1. Advantageous embodiments and further developments as well as a vehicle for carrying out the method emerge from the dependent claims.

[0011] A generic method for issuing instructions to a vehicle user, whereby the time for issuing an instruction is varied depending on whether trigger conditions are met or not, provides that - at least two hint-specific trigger conditions are defined to issue a specific hint; - as soon as a first trigger condition is fulfilled, the respective trigger condition is referred to as the first trigger condition and the at least one further trigger condition is referred to as the second trigger condition and a computing unit determines a probability for the fulfillment of the second trigger condition; - the computing unit compares the probability with a predetermined trigger condition-specific probability threshold; whereby according to the invention: - the computing unit, depending on the comparison, causes the output of the indication at the time the first trigger condition is met and suppresses it at the time the second trigger condition is met; or - the computing unit, depending on the comparison, suppresses the output of the indication at the time the first trigger condition is fulfilled and causes it at the time the second trigger condition is fulfilled; and wherein - the computing unit predicts a temporal progression of the probability, determines from the temporal progression when the probability crosses the predetermined trigger condition-specific probability threshold, and suppresses the output of the indication when the first trigger condition is fulfilled and causes it to be output when the second trigger condition is fulfilled if a crossing duration lasting from the current moment to the crossing time is greater than a specified duration threshold and otherwise causes the output of the indication when the first trigger condition is fulfilled.

[0012] The method according to the invention allows a suitable time for issuing instructions to be determined even more accurately, so that the corresponding information is even more easily perceived by the respective vehicle occupants and distraction in critical driving situations is reduced. Furthermore, the repeated output of the same instruction, which can be perceived as annoying, is avoided. Not only are currently existing boundary conditions taken into account, but by determining the probability of the second trigger condition being met, a future development of the boundary conditions is also taken into account, at least indirectly. The boundary conditions that arise in the future depend on the respective characteristics of the current boundary conditions.Fixed rules can be specified as to how the respective boundary conditions will develop, so that the computing unit is able to determine the probability in a realistic manner.

[0013] A wide variety of information can be provided in the vehicle, including visual, acoustic, and / or haptic information. The information can be directed at the driver and, in addition or alternatively, at other vehicle occupants.

[0014] A wide variety of trigger conditions can be predefined and programmed into the computing unit. A different, trigger-condition-specific probability threshold can be specified for each trigger condition. The respective formula for calculating the probability in accordance with the current boundary conditions can also be selected differently for the various trigger conditions. For some trigger conditions, for example, mathematical equations can be defined that include values measured using vehicle sensors as variables. Other trigger conditions can determine their probability, for example, by retrieving information from the internet, such as the probability of rain, a likely ambient temperature, or the probability of traffic jams.In addition to direct sensor values and information, derived metrics can also be incorporated. For example, a cognitive load factor can be determined from several parameters—for example, fatigue, lane-keeping performance, eye frequency, and external conditions such as rain, snow, or the position of the sun—and incorporated into the evaluation of trigger conditions.

[0015] When comparing the determined probability with the respective trigger condition-specific probability threshold, the probability can be either above or below the probability threshold. Depending on the trigger condition, issuing the hints may be necessary when either the probability threshold is exceeded or below it. This can vary from case to case. Furthermore, depending on the situation, it is advantageous to issue the hint either when the first trigger condition is met or when the second trigger condition is met. Thus, there are four different combinations of when the respective hint should be issued.

[0016] In general, it would also be possible to consider additional trigger conditions in addition to the second trigger condition, such as a third, fourth, or fifth trigger condition. This way, the output of alerts can be deferred for a longer period of time, for example, until the fourth trigger condition is met.

[0017] As described above, the computing unit predicts a temporal progression of the probability, determines from the temporal progression when the probability crosses the predefined trigger condition-specific probability threshold, and suppresses the output of the hint when the first trigger condition is met. It triggers the output of the hint when the second trigger condition is met if the crossing duration from the current moment to the crossing time is greater than a predefined duration threshold, and otherwise triggers the output of the hint when the first trigger condition is met. In other words, the computing unit determines an estimated duration between the times at which the first trigger condition and the second trigger condition will be met. This crossing duration is then compared with the predefined duration threshold.If the duration is too short, the alert will be issued when the first trigger condition is met. If the crossing duration is sufficiently long, the alert will be issued when the second trigger condition is met.

[0018] In this context, we speak of “crossing” the trigger condition-specific probability threshold in order to cover both the case that hints should be issued when the probability threshold is exceeded and the case that hints should be issued when the probability threshold is undershot.

[0019] Depending on the triggering condition, it may be possible to have the computing unit determine the temporal progression of the probability in various ways.

[0020] For example, the trigger condition can be an ambient temperature. A navigation route can be programmed into the vehicle. The vehicle can obtain a weather report from a weather service and derive location-dependent ambient temperatures from this. The computing unit can then compare the navigation route with this weather information. As long as the navigation route leads through areas with, for example, low temperatures, a correspondingly lower probability is determined that a high ambient temperature will prevail. Taking into account the applicable speed limit and the temporal development of the ambient temperatures from the weather report, the vehicle can then determine when there is a sufficiently high probability that a critical ambient temperature, such as 25°C, will be reached.In this case, for example, a message can be given to turn on the vehicle's air conditioning.

[0021] An advantageous development of the method according to the invention provides that if the output of the notification was suppressed upon fulfillment of the first trigger condition, the computing unit only outputs the notification upon fulfillment of the second trigger condition if both the first and second trigger conditions are fulfilled simultaneously, or the computing unit outputs the notification upon fulfillment of the second trigger condition regardless of the degree of fulfillment of the first trigger condition. This allows for even finer adjustment of when the respective notification should be output. For example, it may either be necessary for both trigger conditions to be fulfilled simultaneously, or the fulfillment of the first trigger condition can be evaluated as a trigger to enable the output of the notification upon fulfillment of the second trigger condition, regardless of the degree of fulfillment of the first trigger condition.This also allows the message to be output if the first trigger condition is no longer met when the second trigger condition is met.

[0022] This can be applied analogously to other trigger conditions, so that, for example, the first, second, third and fourth trigger conditions must be met simultaneously for a respective hint to be issued, or it is sufficient that any combination or even just one of these trigger conditions is met after the first trigger condition has been met at least once.

[0023] A further advantageous embodiment of the method according to the invention further provides that at least one triggering condition is formed by: - a time; - a geoposition; - a point of interest; - a temperature; - weather conditions; or - a geofence.

[0024] This makes it possible to take into account many different trigger conditions relevant to the issuing of warnings.

[0025] The vehicle can determine its geoposition, in particular with the help of a navigation unit, such as a navigation system. To do this, the navigation system can receive and evaluate signals sent by a global navigation satellite system. By comparing the current geoposition, i.e. the location of the vehicle, with, for example, a geofence or a point of interest (POI), the issuing of alerts when the vehicle is at or away from a specific location can be specifically controlled. In particular, the time of day can be taken into account as an additional boundary condition. All of the trigger conditions listed can also be considered simultaneously.

[0026] According to a further advantageous embodiment of the method according to the invention, the computing unit determines the probability of fulfilling the second trigger condition using a machine learning model that has been trained for this purpose. The machine learning model can, for example, have been trained, i.e., taught, by the vehicle manufacturer during the development of the vehicle. For this purpose, test campaigns with a wide variety of boundary conditions can be carried out. As a result, the machine learning model gradually learns how high the probability of fulfilling a respective trigger condition is depending on the respective boundary conditions. The machine learning model can then generally be applied to a large number of different users. Using such a machine learning model, a particularly reliable determination of the actual probability is possible.

[0027] Preferably, the machine learning model is continuously trained throughout the vehicle user's operational phase, depending on the vehicle's usage behavior. This makes it possible to adapt the prediction accuracy of the machine learning model to the user's actual usage behavior. Each user will exhibit slightly different usage behavior, so that a generally trained machine learning model will not be able to accurately determine the respective probability for some users. However, this disadvantage can be compensated for by continuously training the respective machine learning model. In this way, the respective machine learning model gradually learns the usage behavior of the vehicle user and is thus able to determine the respective probabilities even more accurately.

[0028] Particularly preferably, the machine learning model is formed by an artificial neural network.

[0029] Groups of different user types can also be classified. Specifically tailored or specifically trained machine learning models can then be maintained for the respective user groups and distributed to the vehicles in a fleet, for example, via a central computing device such as a cloud server. This makes it possible to implement a specifically tailored or trained machine learning model for determining the probability in the respective vehicle's computing units even at the beginning of a vehicle's usage phase. This makes it possible to accurately determine the respective probabilities even at the beginning of the usage phase.

[0030] According to a further advantageous embodiment of the method according to the invention, the vehicle user manually specifies at least one triggering condition. For this purpose, the vehicle user can use any human-machine interface. For example, the manual input can be made via voice command. The vehicle can also be equipped with a touch-sensitive display device through which the vehicle user can enter corresponding commands. This allows the vehicle user to independently create routines that determine which type of notification should be issued when which boundary conditions, and thus which triggering conditions, are met, and at what time. The vehicle user can optionally also specify the respective output modality.

[0031] A further advantageous embodiment of the method according to the invention further provides that at least one notification contains a suggestion for activating or deactivating a vehicle function. For example, the driver can be prompted to turn on the vehicle's air conditioning or heating system depending on the ambient temperature. If a rain shower is imminent, the driver can be prompted to activate a function for automatically closing a sunroof, and the like.

[0032] Preferably, the at least one hint references at least one vehicle function that is content-compatible with the respective vehicle function. This enables the respective vehicle occupant to react appropriately to current boundary conditions in an even more targeted manner. The vehicle can have a wide variety of vehicle functions that are content-compatible with one another. For example, this includes setting a comfortable interior climate. For this purpose, the vehicle has an air conditioning system as well as seat heating and / or seat ventilation. Accordingly, a hint can be issued that suggests how the respective air conditioning system should be set and how the seat heating or seat ventilation should be set. This reduces the risk that the respective vehicle occupant will not use the seat heating function and / or seat ventilation function, for example.

[0033] In a vehicle comprising output means for visually, acoustically and / or haptically outputting instructions to vehicle users, detection means for detecting information and a computing unit for controlling the output means depending on processing of the information, the output means, the detection means and the computing unit are configured according to the invention to carry out a method described above. The vehicle can be any road vehicle such as a car, truck, van, bus or the like. It could generally also be a rail vehicle, watercraft or aircraft. The vehicle can have, for example, loudspeakers, display devices, in particular touch-sensitive display devices, actuators for imparting vibrations and the like as output means.The vehicle can be equipped with a wide variety of sensors as detection means, such as temperature sensors, wheel speed sensors, acceleration sensors, brightness sensors, and the like. Furthermore, communication means such as a telecommunications unit can also be referred to as detection means, since such communication means can retrieve information from sources external to the vehicle, such as a service provider accessible via the Internet. The processing unit can be, for example, a central on-board computer, the control unit of a vehicle subsystem, or the like. Thus, several control units can also interact to execute the method steps provided by the processing unit.

[0034] Further advantageous embodiments of the method according to the invention for outputting instructions to a vehicle user and of a vehicle according to the invention also emerge from the exemplary embodiments which are described in more detail below with reference to the figures.

[0035] Showing: Fig. 1 a schematic representation of the timing of the notifications to a vehicle user depending on the fulfillment of two trigger conditions; Fig. 2 a table showing possible combinations of when an indication should be issued depending on a comparison between a probability of fulfilling a second trigger condition and a trigger condition-specific probability threshold; Fig. 3 a schematic representation of a first example of two situations requiring the output or suppression of cues; and Fig. 4 a schematic representation of a second example of three situations requiring the output or suppression of cues.

[0036] Fig. 1 shows on several timelines a respective time t1, t2 at which a first trigger condition BED1 and a second trigger condition BED2 are fulfilled, as well as when the output of an indication to be output when a respective trigger condition BED1, BED2 is fulfilled takes place.

[0037] Fig. 1 shows in the Fig. 1a), Fig. 1b) and Fig. 1c) Three different possibilities. A solid black circle indicates when a respective condition BED1, BED2 is met, and an empty circle indicates when the respective condition BED1, BED2 is not met. A small "X" marks the time t1, t2 at which the respective hint is issued.

[0038] In Fig. 1a) The first trigger condition BED1 is met at time t1, and the second trigger condition BED2 is not met. At time t2, both trigger conditions BED1 and BED2 are met. In the example shown here, the notification is issued as soon as the first trigger condition BED1 is met at time t1.

[0039] In Fig. In 1b), however, the notification is only issued when the second trigger condition BED2 is met at time t2. This requires that at least the first trigger condition BED1 be met simultaneously. There could also be additional trigger conditions not shown in detail, which would then also have to be met.

[0040] In Fig. 1c) it is sufficient if only the second trigger condition BED2 is fulfilled at time t2 for the hint to be issued.

[0041] Fig. 2 shows in a table the possible variants of when the respective notification should be issued in the vehicle according to a method according to the invention. First, the said trigger conditions BED1, BED2 are defined for the output of a specific notification. As soon as the first trigger condition BED1 is met, a probability P for the fulfillment of the second trigger condition BED2 is determined. A computing unit of a vehicle of a vehicle user determines the probability P and then compares it with a predetermined trigger condition-specific probability threshold value PSW. Depending on the comparison, the computing unit then causes the notification to be output either at time t1 or at time t2 and correspondingly suppresses the output of the notification at the other time t2, t1.

[0042] This shows Fig. 2 The first column of the table shows the respective results of the comparison of the probability P of a respective trigger condition BED with the associated trigger condition-specific probability threshold PSW. The second column of the table shows which trigger condition the notification should be issued when met. According to the first row of the table, the probability P exceeds the trigger condition-specific probability threshold PSW and the notification is issued when the first trigger condition BED1 is met. According to the second row, however, in this case the notification is only issued when the second trigger condition BED2 is met. The second and third rows of the table describe corresponding cases in which the probability P falls below the trigger condition-specific probability threshold PSW.

[0043] Fig. Figure 3 shows a possible application of the method according to the invention. It shows a vehicle 1 according to the invention parked in a garage 2. The garage 2 is separated from the surroundings 4 by a garage door 3. The temperature inside the garage is 4°C. Fig. 3a) shows the case for Monday, where at 6 a.m. the outside temperature is -4°C. In Fig. 3b) shows the case for Tuesday, when the outside temperature at 6 a.m. is 6°C.

[0044] By observing the vehicle user's usage behavior, a machine learning model running on a computing unit (not shown in detail) of vehicle 1 deduces that the vehicle user typically turns on their seat heating at 6 a.m., so it would be appropriate to issue appropriate notifications to the vehicle user at this time to turn on the seat heating should the vehicle user forget to turn on the seat heating at 6 a.m., or shortly before or after. At the same time, the vehicle user has defined a rule that the seat heating should be turned on automatically when the outside temperature is below 0°C.

[0045] Vehicle 1 can determine, with the aid of its environmental sensors and possibly taking into account information derived from a navigation unit, that it is in garage 2. This information can also be stored when vehicle 1 is switched off or the processing unit is deactivated. The processing unit can receive a weather report from a weather service. The weather forecast was current at the time of receipt. For example, it can be deduced from the weather forecast that it will be -4°C on Monday at 6:00 a.m., 6°C on Tuesday, and -2°C on Wednesday. Fig. 3a) the notification to switch on the seat heating should not be issued, since a corresponding automation routine is implemented in vehicle 1, which automatically switches on the seat heating when the temperature falls below 0°C. Fig. In the example shown in 3b), however, the warning should be issued that the seat heating is not automatically switched on because the outside temperature is greater than 0°C. Since it is still cold, the vehicle user might still prefer to switch on the seat heating.

[0046] In Fig. 4 shows a journey of the vehicle 1 along a navigation route 5. Along the navigation route 5 there is a geofence 6. In Fig. In the example shown in Figure 4a), navigation route 5 passes through geofence 6, with the estimated time of arrival (ETA) being reached in two minutes. Trigger conditions could be defined that cause alerts to be issued during the stay in geofence 6. The expected time of arrival (ETA) can also be used as a trigger condition.

[0047] The implementation could also be different. For example, arriving at the destination could represent the fulfillment of a trigger condition. The expected arrival time ETA could then correspond to a crossing duration at which a corresponding probability P exceeds or falls below the respective trigger condition-specific probability threshold PSW, i.e., crosses. For example, 10 minutes could be specified as the specified duration threshold. This makes it possible to issue notifications in the Fig. 4a) is not necessary, since the crossing duration of 2 minutes is less than the duration threshold of 10 minutes.

[0048] In Fig. 4c), however, the expected arrival time ETA will be reached in 20 minutes, so that the crossing duration is greater than the specified duration threshold, so that warnings should be issued.

[0049] In Fig.4b), however, the navigation route 5 leads past the geofence 6, so that generally no instructions should be given.

Claims

[1] Method for issuing instructions to a vehicle user, wherein the time for issuing an instruction is varied depending on whether trigger conditions (BED1, BED2) are met or not met, wherein - at least two hint-specific trigger conditions (BED1, BED2) are defined to issue a specific hint; - as soon as a first triggering condition (BED1) is fulfilled, the respective triggering condition (BED1) is referred to as the first triggering condition (BED1) and the at least one further triggering condition (BED2) is referred to as the second triggering condition (BED2) and a computing unit determines a probability (P) for the fulfillment of the second triggering condition (BED2); - the computing unit compares the probability (P) with a predetermined trigger condition-specific probability threshold (PSW); characterized by , that - the computing unit, depending on the comparison, causes the output of the indication at the time (t1) when the first trigger condition (BED1) is fulfilled and suppresses it at the time (t2) when the second trigger condition (BED2) is fulfilled; or - the computing unit, depending on the comparison, suppresses the output of the indication at the time (t1) when the first trigger condition (BED1) is fulfilled and causes it at the time (t2) when the second trigger condition (BED2) is fulfilled; and wherein - the computing unit predicts a temporal progression of the probability (P), determines from the temporal progression when the probability (P) crosses the predefined trigger condition-specific probability threshold (PSW), and suppresses the output of the indication when the first trigger condition (BED1) is fulfilled and causes it to be output when the second trigger condition (BED2) is fulfilled if a crossing duration lasting from the current moment to the crossing time is greater than a specified duration threshold and otherwise causes the output of the indication when the first trigger condition (BED1) is fulfilled. [2] Method according to claim 1, characterized bythat if the output of the hint upon fulfillment of the first trigger condition (BED1) was suppressed, the computing unit only causes the output of the hint upon fulfillment of the second trigger condition (BED2) if both the first and the second trigger condition (BED1, BED2) are fulfilled at the same time, or the computing unit causes the output of the hint upon fulfillment of the second trigger condition (BED2) regardless of the degree of fulfillment of the first trigger condition (BED1). [3] Method according to claim 1 or 2, characterized by that at least one trigger condition (BED1, BED2) is formed by: - a time; - a geoposition; - a point of interest; - a temperature; - weather conditions; or - a geofence (6). [4] Method according to one of claims 1 to 3, characterized bythat the computing unit determines the probability (P) of fulfilling the second trigger condition (BED2) by means of a machine learning model learned for this purpose. [5] Method according to claim 4, characterized by that the machine learning model is continuously trained throughout the deployment phase of the vehicle (1) of the vehicle user depending on the usage behavior of the vehicle (1). [6] Method according to one of claims 1 to 5, characterized by that the vehicle user manually specifies at least one trigger condition (BED1, BED2). [7] Method according to one of claims 1 to 6, characterized by that at least one note contains a suggestion to activate or deactivate a vehicle function. [8] Method according to claim 7, characterized by that the at least one reference references at least one vehicle function that is compatible in content with the respective vehicle function. [9] Vehicle (1), comprising output means for the visual, acoustic and / or haptic output of instructions to vehicle users, recording means for recording information and a computing unit for controlling the output means depending on a processing of the information, characterized by that the output means, the detection means and the computing unit are arranged to carry out a method according to one of claims 1 to 8.

Images