Method for operating a driver assistance device for a motor vehicle, corresponding driver assistance device and computer program product

Abstract

The invention relates to a method for operating a driver assistance device for a motor vehicle (1), wherein environmental data describing the environment of the motor vehicle are recorded at least temporarily by means of an environment detection device of the motor vehicle and, using the environment data, autonomous driving operation of the motor vehicle (1) is carried out at least temporarily without any operating action by occupants of the motor vehicle (1) at a control element provided and designed for influencing the driving operation.The invention provides for the determination of a confidence level describing the driver assistance system's confidence in performing autonomous driving, as well as a confidence level describing the occupants' confidence in performing autonomous driving. When the confidence level exceeds a certain threshold, and when it falls below a certain threshold, a message aimed at increasing the confidence level is generated and displayed to the occupants, depending on the environmental data. The determination of the confidence level is based on behavioral recognition of the occupants of the motor vehicle (1). The invention further relates to a driver assistance system for a motor vehicle and a computer program.

Description

The invention relates to a method for operating a driver assistance system for a motor vehicle, wherein environmental data describing the environment of the motor vehicle is acquired at least temporarily by means of an environment sensing device of the motor vehicle, and autonomous driving operation of the motor vehicle is carried out at least temporarily using the environment data without any operating action by the occupants of the motor vehicle at a control element provided and designed for influencing the driving operation, wherein a confidence value describing the confidence of the driver assistance system for carrying out the autonomous driving operation and a confidence value describing the confidence of the occupants in carrying out the autonomous driving operation are determined, and if a confidence value exceeds a confidence threshold, in particular only or exclusively then and / or already then,and, if the confidence level falls below a certain threshold, in particular only or exclusively then and / or already then, a message aimed at increasing the confidence level is generated and issued to the occupants depending on the environmental data, wherein the determination of the confidence level is based on behavioral recognition of the occupants of the motor vehicle. The invention further relates to a driver assistance device for a motor vehicle and a computer program product. For example, the prior art includes the publication DE 10 2019 006 685 A1. This describes a method for operating a vehicle in an automated driving mode that requires no user action, which can be deactivated by a deactivation action by the driver of the vehicle, and in which driving situations in the vicinity of the vehicle are detected by means of at least one environment detection device and a warning is issued to the driver if a critical driving situation exists. The system envisages that, during automated driving operation, at least one learning phase will record driving situations in which the driver has deactivated automated driving. These recorded driving situations will be stored in a memory as subjectively critical driving situations. Furthermore, during a regular operating phase of automated driving, the system will compare the extent to which a currently recorded driving situation corresponds to the stored subjectively critical driving situations, and if there is sufficient similarity, a warning will be issued to the driver. German patent application DE 10 2021 003 073 B3 concerns a method for increasing the proportion of automated driving in a vehicle with at least partial automation, in which the conditions for automated driving are monitored. In driving situations where the number of manual takeovers by the driver is reduced in order to utilize the advantages of automated driving optimally designed for traffic flow, road safety, and energy efficiency, a model predicts the termination of automated driving by manual control when conditions for automated driving are met. In the event of a predicted takeover by the driver, information is output to the driver, informing them that the automated driving system is reliably controlling the driving situation. Further systems for autonomous driving of a motor vehicle are known from the publications US 2020 / 0223450 A1 and DE 10 2023 104 508 A1. The object of the invention is to propose a method for operating a driver assistance system for a motor vehicle which has advantages over known methods, in particular achieving a particularly high acceptance of autonomous driving operation among the occupants of the motor vehicle. According to the invention, this is achieved by a method for operating a driver assistance system for a motor vehicle with the features of claim 1. It is provided that the message is only generated and output if a decrease in the confidence level is detected simultaneously with a decrease in the confidence level. In principle, it is intended that a confidence level describing the confidence of the driver assistance system in carrying out autonomous driving operations, as well as a confidence level describing the confidence of the occupants in carrying out autonomous driving operations, are determined, and if the confidence level exceeds a confidence threshold, in particular only or exclusively then and / or already then, and if the confidence level falls below a confidence threshold, in particular only or exclusively then and / or already then, a message aimed at increasing the confidence level is generated and issued to the occupants depending on the environmental data, whereby the determination of the confidence level is based on behavioral recognition of the behavior of the vehicle occupants. Advantageous embodiments with expedient further developments of the invention are specified in the dependent claims. It should be noted that the exemplary embodiments described in the description are not limiting; rather, any variations of the features disclosed in the description, the claims, and the figures are possible. The method serves to operate the driver assistance system. The driver assistance system, or a control unit of the driver assistance system used to carry out the method, is preferably a component of the motor vehicle, but can of course also exist separately from it, particularly until the driver assistance system or the control unit is mounted on or in the motor vehicle. The driver assistance system serves, at least temporarily, to carry out driving operations of the motor vehicle, which in this case are autonomous. Autonomous driving operation is understood to mean that the driver assistance system independently takes over longitudinal and lateral control of the motor vehicle, at least temporarily. Longitudinal guidance refers in particular to setting the vehicle's speed, for example, by appropriately controlling the vehicle's drive system and / or braking system. The drive system serves to propel the vehicle, thus providing the drive torque for propelling the vehicle. To provide the drive torque, the drive system preferably has at least one drive unit, which can be configured as an internal combustion engine or an electric traction motor. The braking system, on the other hand, serves to brake or decelerate the vehicle. It includes, in particular, a service brake, which exerts a braking force on one or more wheels of the vehicle to slow it down. Assistance to the user in performing driving operations occurs, for example, within the framework of assisted driving according to SAE Level 1. At least partially autonomous driving operations can occur, in particular, within the framework of partial automation according to SAE Level 2. According to the invention, the autonomous driving operation of the motor vehicle takes place within the framework of conditional automation according to SAE Level 3, high automation according to SAE Level 4, or full automation according to SAE Level 5. The SAE Levels are also known as levels, so the aforementioned levels can be designated as L1 to L5. Whenever autonomous driving is mentioned in this description, it always refers to autonomous driving according to the above definition, in particular according to SAE Level 3 or higher, and specifically not to assisted driving or only partially autonomous driving according to SAE Levels 1 and 2. The vehicle, and in particular its driver assistance system, is equipped with an environmental sensing system that serves to detect the vehicle's surroundings. When the environmental sensing system detects the surroundings, it generates environmental data that describes the vehicle's environment. To detect the surroundings and provide this environmental data, the environmental sensing system has one or more environmental sensors. Examples of sensors used as environmental sensors include: radar sensor, sonar sensor, lidar sensor, image sensor or camera, and sound sensor or microphone. The radar sensor uses electromagnetic waves, the sonar sensor uses sound waves, and the lidar sensor uses light to detect the environment, particularly to identify obstacles. The sonar sensor, for example, is designed as an ultrasonic sensor and therefore uses sound waves in the ultrasonic range. The radar sensor, sonar sensor, and lidar sensor are all reflectivity sensors, meaning they are active sensors that emit an output signal into the environment and receive the reflected output signal as an input signal. The sound sensor, or microphone, is used to detect sound waves present in the environment. The driver assistance system is designed to perform autonomous driving at least temporarily based on environmental data. This means that environmental data is evaluated, particularly regarding one or more obstacles in the vicinity of the vehicle. Autonomous driving is based on obstacle data that describes the obstacles present in the vehicle's environment. Where the plural "obstacles" is used in this description, it is understood that the number of obstacles actually present depends on the situation. The number of obstacles present in the vehicle's environment, or the number of obstacles described by the obstacle data, is arbitrary. There may be no obstacle at all, only one obstacle, or there may be several obstacles present or described by the obstacle data.Obstacle data is acquired during obstacle detection. This process involves, for example, sensor fusion, in which environmental data from various sensors is combined. During obstacle detection, the environmental data provided by the sensor is analyzed and examined for obstacles. If obstacles are detected, the obstacle data describing them is generated and provided. For each obstacle, the obstacle data contains at least one parameter that describes the corresponding obstacle.Preferably, one or more of the following parameters are used as obstacle parameters: position of the obstacle, orientation of the obstacle, outline of the obstacle, direction of movement of the obstacle, speed of movement of the obstacle, direction of acceleration of the obstacle, acceleration of the obstacle and type of obstacle. The type of obstacle is determined, for example, through its classification. The type of obstacle preferably describes whether it is stationary or moving. A stationary obstacle is, for example, a structure, particularly a building, or vegetation, particularly a tree or the like. A moving obstacle is understood to be, for example, a road user other than the motor vehicle, i.e., in particular, a pedestrian or another vehicle, preferably another motor vehicle, a bicycle, or the like. Preferably, the type of obstacle also describes a type or class of the other motor vehicle; the EU vehicle class is preferably used as a basis for this. The obstacles can also be referred to as objects. Autonomous driving of a motor vehicle refers specifically to driving mode during which the user does not interact with the controls. The user is one of the occupants, for example, the driver of the vehicle, who temporarily takes over control of the vehicle outside of autonomous driving mode. The controls are designed and configured to influence the driving mode. Therefore, when the user interacts with the controls, they are influencing the driving mode. The control element is, for example, a steering wheel, an accelerator pedal, a brake pedal, or the like. Specifically, the control element is designed and intended to influence the direction and / or speed of the motor vehicle. The act of operating the vehicle encompasses any interaction between the user and the control element; the user does not necessarily have to actually influence the vehicle's operation by using the control element, it is sufficient that the possibility of doing so exists. The act of operating the vehicle is, for example, simply touching the control element. Where the term "occupants" is used in the plural within this description, this does not imply a quantification. The number of occupants is arbitrary; there may be only one occupant or there may be several occupants on board the motor vehicle. Autonomous vehicles require both societal and occupant acceptance to prevail over other vehicles. The safe operation of the vehicle is typically considered a crucial factor. This includes, in particular, the occupants' sense of security during autonomous driving. Driver monitoring systems (DMS) are already widespread; these systems measure and assess the attention and gaze direction of the user, especially the driver. Such systems are also important for autonomous driving, as it is permissible for occupants to engage in activities unrelated to driving during autonomous operation.For example, some occupants focus on work during such non-driving-related activities, some on their family, and others on the scenery. The aim is to create a procedure that maintains a consistently high level of occupant confidence during autonomous driving. To this end, the confidence level and the confidence measure are determined. The confidence level describes the confidence of the driver assistance system itself, which is used to perform autonomous driving. In other words, the confidence level describes the probability that a sufficiently high level of vehicle safety is ensured, or can be ensured, by the driver assistance system during autonomous driving. For example, the autonomous driving mode is designed to end as soon as the confidence level falls below the confidence threshold. Specifically, a warning signal is generated when the confidence level falls below the threshold, alerting the occupants or user of the vehicle that autonomous driving is being terminated and that they must take over manual control. After a certain period following the warning signal, autonomous driving is deactivated, and the driver assistance system assumes that the user is manually controlling the vehicle. The confidence level, on the other hand, describes the occupants' trust in the autonomous driving system and its execution. The higher the confidence level, the more the occupants trust the driver assistance system to perform autonomous driving safely and reliably. Conversely, if the occupants doubt the driver assistance system's ability to perform autonomous driving safely, the confidence level is lower. The confidence level thus also reflects the level of comfort that the user or occupants of the vehicle attribute to the vehicle. The confidence level is determined based on behavioral recognition. Behavioral recognition is designed and configured to evaluate the behavior of the occupants and generate and provide a behavioral metric based on this behavior. The confidence level depends significantly on this behavioral metric. Behavioral recognition assesses, for example, facial expressions and / or gestures of the occupants, their posture, interactions between occupants, and / or the performance of activities unrelated to driving. The system is designed to generate and display a message to the user based on the confidence level and the confidence threshold. Specifically, the message is generated when the confidence level exceeds the confidence threshold. This means the message is generated and displayed when the driver assistance system assumes the autonomous driving operation is sufficiently safe. Additionally, the confidence level is considered during message generation and display. The message is generated and displayed if the confidence level falls below the confidence threshold. For each of the aforementioned conditions, it is preferable that the message is only generated and output if the respective condition is met. In any case, both conditions must be fulfilled. Generation and output are therefore omitted if either condition is not met. Thus, the message is only generated and output to the user if this is the case, i.e., if both the confidence level exceeds the confidence threshold and the confidence level falls below the confidence threshold. The message is generated, for example, using artificial intelligence, specifically an artificial neural network. In particular, a generative neural network is employed, which receives at least the environmental data as input. Additionally or alternatively, the neural network also considers the confidence level and / or the trust level as inputs. The message is generated by the neural network, or the network provides the message as an output for subsequent output. The message content is determined based on the surrounding data. Specifically, the generated message describes the environment based on the environmental data collected by the vehicle's environmental sensing system. This provides the vehicle's occupants with a plausible explanation of the basis for autonomous driving. This approach aims to increase confidence in the system. The described procedure is used particularly for traffic situations that the vehicle can easily handle during autonomous driving, and indeed, only in such situations. For example, this approach is used in the following situation: The occupants are engaged in an activity unrelated to driving, such as consuming entertainment media and paying no attention to the vehicle's operation. Eventually, one of the occupants happens to look up and notices the vehicle's surroundings. If, for instance, they see the end of a traffic jam ahead, doubts about the autonomous driving operation may arise, and the confidence level decreases. This is registered by the driver assistance system, and if the confidence level is greater than the confidence threshold, messages are generated and displayed to the occupants, aimed at increasing the confidence level. The message reassures the occupant that the vehicle is safe and that they can return to their unrelated activity.For example, the driver assistance system may also suggest non-driving-related activities to the occupant(s). This approach helps maintain trust in autonomous driving. The message can be delivered in any way imaginable. For example, it can be delivered visually and / or audibly. In the former case, the message is visual; in the latter, it is audible. A combination of both message types is particularly preferred, resulting in an audiovisual message. The message can be delivered, for example, via a screen and / or a loudspeaker located inside the vehicle. The invention provides that the message is generated and output only if a decrease in the confidence level is detected simultaneously with a decrease in the confidence level. In this case, it is assumed that the decrease in the confidence level is caused by a complex traffic situation, which also leads to a decrease in the confidence level. Accordingly, the message is generated and output only if both the confidence level and the confidence level decrease. However, it remains the case that the message is only generated and output if the confidence level is greater than the confidence threshold. If the confidence level falls below the confidence threshold, the aforementioned warning signal is preferably generated and the autonomous driving operation is terminated.The described procedure achieves a particularly high level of plausibility for the autonomous driving operation for the occupants of the vehicle. A further development of the invention provides that an attention parameter describing the occupants' attention directed towards the autonomous driving operation of the motor vehicle and / or an attention target describing the direction of the occupants' attention are determined, wherein the attention parameter and / or the attention target are taken into account when generating the message, or the generation takes place, in particular only or exclusively when and / or already when the attention parameter exceeds an attention threshold. The attention parameter describes the occupants' attention directed towards the driving operation of the motor vehicle or their attention directed towards the surroundings. For example, the attention parameter is greater the longer the occupants observe their surroundings per unit of time, in particular by looking out of a window of the motor vehicle. In addition to the level of attention, it may be necessary to determine the attention target. This describes a goal of the occupants' attention. Preferably, the level of attention and / or the attention target are taken into account when generating the message. However, it may also be possible to generate the message only if the level of attention exceeds the attention threshold. Thus, in addition to the level of attention falling below the confidence threshold, the level of attention also needs to exceed the attention threshold. This ensures that the message is not triggered by a misinterpretation of the confidence level, for example, if the change in the confidence level is not due to vehicle operation. Overall, this achieves a high degree of accuracy in generating and outputting the message.A further development of the invention provides that behavior detection is carried out using at least one interior sensor, which detects the occupants located in the interior of the motor vehicle. The interior sensor is directed towards the interior of the motor vehicle in which the occupants are located. The interior sensor is, for example, an image sensor, a sensor for physiological parameters, a sound sensor, or the like. The image sensor, for example, detects user movements. If these movements exceed a certain speed and / or occur at a frequency greater than a threshold, this indicates increased restlessness among the occupants. Accordingly, the detected movements are incorporated into the behavior detection system or taken into account by it. For example, it is checked whether one of the occupants moves a hand towards the steering wheel.This behavior clearly indicates an intention to take over driving the vehicle. The physiological parameter sensor is designed and configured to record at least one physiological parameter of the occupants. In this respect, the physiological parameter sensor serves, for example, to determine body temperature, skin color, in particular changes in skin color, or similar parameters. Different sensors can also be used, which are evaluated using artificial intelligence, preferably an artificial neural network, to determine the confidence level from the sensor data they provide within the framework of behavior recognition. The confidence level is not merely derived indirectly, but is measured using at least one indoor sensor. This achieves a high degree of accuracy. A further development of the invention provides that the message is generated based on environmental data and sensor data from at least one interior sensor. The message thus depends not only on environmental data but also, at least additionally, on the sensor data from the interior sensor. This allows the current state of the occupants to be addressed particularly effectively using the message, thereby building trust in autonomous driving operation. A further development of the invention provides that the activities of the occupants are determined from the sensor data and that these activities are used to generate the message. These activities are understood to be the activities the occupants are currently engaged in. The content of the message is based on these activities; for example, the occupants are reassured in this way that they can continue with their activities because the autonomous driving operation is safe and reliable. This also serves to establish trust in the functionality of the driver assistance system. A further development of the invention provides that, in addition to generating and outputting the message, an operating parameter of the vehicle that influences driving behavior is adjusted. For example, the vehicle's speed is adjusted, and in particular, it is reduced more significantly the lower the confidence level. Thus, the vehicle's driving behavior is also adapted to the moment the confidence level falls below the threshold, in order to increase the occupants' confidence again. In combination with generating and outputting the message, this is particularly effective from a psychological perspective. A further development of the invention provides that, based on environmental and sensor data and using configuration data, a notification data set is selected from several notification data sets stored in a data store, and the message is compiled using a notification rule stored in the selected notification data set. The data store contains any number of notification data sets. The notification data set is selected from this number of notification data sets using the environmental data and the sensor data from the indoor sensor. Configuration data is used to define the selection process.For example, the configuration data represents parameters of an artificial neural network, to which the environmental data and sensor data are fed as input and which provides the notification data sets as output. Each notification record contains a notification rule. Accordingly, the selected marker record also contains such a notification rule. The notification rule defines the composition of the message; for example, it specifies which information is included when the message is generated. This approach enables particularly targeted message generation. A further development of the invention provides that the notification rule includes a provision specifying whether or not the following data should be output: current and / or predicted driving situation, current and / or predicted driving maneuver, expected autonomous driving time, entertainment and / or information media, environmental conditions, and suggested action. Depending on the notification rule, the aforementioned data is therefore included in the message or not when it is generated. The driving situation describes the current and / or future traffic situation in which the vehicle is or will be located. A driving maneuver is a maneuver currently or in the future performed by the vehicle during autonomous driving operation. The expected autonomous driving time describes how long the vehicle is expected to continue operating autonomously before, for example, the confidence level is expected to fall below the confidence threshold. Entertainment and / or information media describe any media aimed at entertaining and / or informing the occupants. Such media can include, for example, films, music, news, or the like. Environmental conditions describe at least one situation in the vehicle's environment, such as a traffic situation.The activity suggestion includes a proposal for a non-driving-related activity that can be carried out by the occupants. In this respect, an arbitrary compilation of the message is achieved, thus making a particularly significant increase in the level of confidence attainable. A further development of the invention provides that a history of the confidence level is transmitted together with the generated message to an external computing unit and used to adjust notification data sets stored there, with the adjusted notification data sets being transmitted to the driver assistance system. The history of the confidence level describes, in particular, the progression from the moment the message is issued, so that it can be determined from the history whether the message is issued leads to a change in the confidence level. Accordingly, the success of the implemented measure can be inferred from the trend of the confidence interval. This trend, along with the generated and output message, is transmitted to the computing unit. The computing unit is understood to be a computing device located at a distance from the vehicle, in particular a stationary computing device. Preferably, the computing unit has a computing power greater than that of the driver assistance system or its control unit. The computing unit uses the transmitted confidence level and message history to determine the message's success. Depending on the confidence level history and the message, the notification data records are adjusted. These adjusted notification data records are preferentially transmitted to the driver assistance systems, potentially replacing any previously existing notification data records. This approach ensures a gradual adaptation of the generated and output message to the needs of the vehicle's occupants. The invention further relates to a driver assistance arrangement for a motor vehicle, in particular for carrying out the method according to the explanations within the scope of this description, wherein the driver assistance device is provided and designed to acquire environmental data describing the environment of the motor vehicle at least temporarily by means of an environment detection device of the motor vehicle and to carry out autonomous driving operation of the motor vehicle at least temporarily without any operating action by occupants of the motor vehicle at a control element provided and designed for influencing the driving operation. The driver assistance system is also designed and configured to determine a confidence level describing the driver assistance system's confidence in performing autonomous driving operations, as well as a confidence level describing the occupants' confidence in performing autonomous driving operations. When the confidence level exceeds a certain threshold, particularly only or exclusively when and / or already when, and when the confidence level falls below a certain threshold, particularly only or exclusively when and / or already when, a message aimed at increasing the confidence level is generated and output to the occupants, depending on the environmental data. The determination of the confidence level is based on behavioral recognition of the vehicle occupants' behavior.The driver assistance system is also designed and configured to generate and output the message only when a decrease in the confidence level is detected simultaneously with a decrease in the confidence level. The advantages of such a design of the driver assistance system and such a procedure have already been pointed out. Both the driver assistance system and the procedure for operating it can be further developed as explained in the description, and reference is made to that description in this regard. Furthermore, the invention relates to a computer program product comprising commands that cause the driver assistance system to execute the described method as outlined in this description. For the advantages and possible further developments, reference is made to the description in its entirety. The features and combinations of features described in the description, in particular those described in the following figure description and / or shown in the figures, are not only usable in the combinations specified, but also in other combinations or individually, without departing from the scope of the invention, in particular the scope of the claims. Thus, embodiments that are not explicitly shown or explained in the description and / or the figures, but which emerge from or can be derived from the explained embodiments, particularly within the scope of the claims, are also to be considered as encompassed by the invention. The invention is explained in more detail below with reference to the exemplary embodiments shown in the drawing, without limiting the invention. Figure 1, the only figure, shows a schematic representation of a traffic situation, which serves to illustrate a method for operating a driver assistance system for a motor vehicle. Figure 1 shows a schematic representation of a traffic situation for a motor vehicle 1. The motor vehicle 1 is traveling on a roadway 2, in this example a roadway 2 with several lanes 3. Other road users 4, also motor vehicles, are present on the roadway 2. Also shown is an oncoming roadway 5 with lanes 6, on which additional road users 7 are located. The motor vehicle 1 has an environmental detection system that detects the area surrounding the motor vehicle 1. In this embodiment, both road users 4 and road users 7 are present in the environment, so they are recognized and evaluated as obstacles. The vehicle 1, or rather a driver assistance system of the vehicle 1, is operating autonomously. This means that the vehicle 1 is operating without any input from the vehicle's occupants at a control element that would influence the driving process. It is planned to determine, during the autonomous driving operation of the vehicle 1, both a confidence level and a trust level. The confidence level describes the confidence of the driver assistance system, while the trust level describes the occupants' confidence in the proper execution of the autonomous driving operation.If the confidence level exceeds a confidence threshold and simultaneously falls below a confidence threshold, a message is generated for the user, based on environmental data acquired by an environmental sensing device in vehicle 1. The generated message is then output, for example, by displaying it on a screen in vehicle 1 and / or using a loudspeaker. The message is preferably generated using an artificial neural network, which preferably executes a Large Language Model (LLM). The message is generated in such a way as to increase confidence. The described approach ensures that the occupants of vehicle 1 perceive autonomous driving as particularly reliable and safe, and that any uncertainty that may arise is quickly and effectively resolved by explaining the current traffic situation and / or the driving maneuvers currently being performed. Preferably, a history of the confidence level recorded after the message is sent, as well as the message itself, are transmitted to an external computing unit 8. There, the success of the message is evaluated, and if necessary, the rule governing message output is adjusted. This leads to an improvement in the message over time. REFERENCE MARK LIST: 1 Motor vehicle 2 Roadway 3 Lane 4 Road user 5 Oncoming lane 6 Lane 7 Road user 8 Computing device

Claims

Method for operating a driver assistance device for a motor vehicle (1), wherein environmental data describing the environment of the motor vehicle are recorded at least temporarily by means of an environment detection device of the motor vehicle and, using the environment data, autonomous driving operation of the motor vehicle (1) is carried out at least temporarily without any operating action by occupants of the motor vehicle (1) at an operating element designed and configured for influencing the driving operation,wherein a confidence value describing the confidence of the driver assistance system in carrying out autonomous driving operation and a confidence value describing the confidence of the occupants in carrying out autonomous driving operation are determined, and if the confidence value exceeds a confidence threshold or falls below a confidence threshold, a message aimed at increasing the confidence value is generated and output to the occupants depending on the environmental data, wherein the determination of the confidence value is based on behavioral recognition of the behavior of the occupants of the motor vehicle (1), characterized in that the message is only generated and output if a decrease in the confidence value is detected simultaneously with a decrease in the confidence value. Method according to claim 1, characterized in that an attention quantity describing the attention of the occupants directed towards the autonomous driving operation of the motor vehicle (1) and / or an attention target describing the direction of the attention of the occupants are determined, wherein the attention quantity and / or the attention target is / are taken into account when generating the message or the generation takes place when the attention quantity exceeds an attention threshold. Method according to one of the preceding claims, characterized in that the behavior detection is carried out using at least one interior sensor by means of which the occupants located in an interior of the motor vehicle (1) are detected. Method according to claim 3, characterized in that an image sensor and / or a physiological parameter sensor detecting at least one physiological state parameter of the occupants is used as the interior sensor. Method according to claim 3 or 4, characterized in that the message is generated depending on the environmental data and sensor data of the at least one indoor sensor. Method according to one of the preceding claims, characterized in that, in addition to generating and outputting the message, an operating parameter of the motor vehicle (1) influencing the driving operation is adjusted. The method according to claim 5, characterized in that, based on the environmental data and the sensor data, a notification data set is selected from several notification data sets stored in a data storage device using configuration data, and the message is compiled using a notification rule stored in the selected notification data set. Driver assistance device for a motor vehicle (1), in particular for carrying out the method according to one or more of the preceding claims, wherein the driver assistance device is provided and configured to acquire, at least temporarily, environmental data describing the environment of the motor vehicle (1) by means of an environment detection device of the motor vehicle (1) and, using the environment data, to carry out, at least temporarily, autonomous driving of the motor vehicle (1) without any operating action by occupants of the motor vehicle (1) at a control element provided and configured for influencing the driving operation, wherein the driver assistance device is also provided and configured toto determine a confidence level describing the confidence of the driver assistance system in carrying out autonomous driving operation, as well as a confidence level describing the confidence of the occupants in carrying out autonomous driving operation, and, depending on the environmental data, to generate and output a message to the occupants aimed at increasing the confidence level when the confidence level exceeds a confidence threshold or when the confidence level falls below a confidence threshold, wherein the determination of the confidence level is based on behavioral recognition of the behavior of the occupants of the motor vehicle (1), characterized in that the driver assistance system is also designed and configured to generate and output the message only if a decrease in the confidence level is detected simultaneously with a decrease in the confidence level. Computer program product comprising commands that cause the driver assistance device according to claim 8 to execute the method according to one or more of claims 1 to 7.

Images