DE102022126736A1 - Procedure for operating an emergency information system, motor vehicle and emergency information system - Google Patents

Abstract

Method for operating an emergency information system (23) comprising at least one motor vehicle (1), wherein the motor vehicle (1) has an interior (2) for at least one occupant, at least one vehicle sensor recording occupant impact sensor data describing the movement of at least one of the at least one occupant and/or forces acting on the occupant, and a storage means (20), wherein in the storage means (20) for generating an emergency information data set - at least health information describing one health characteristic of the occupant and - the occupant impact sensor data related to at least a predefined retention period are stored and, if an accident signal indicating an accident is present, the emergency information data set is released for retrieval by and/or transmission to an emergency service.

Description

The invention relates to a method for operating an emergency information system comprising at least one motor vehicle, wherein the motor vehicle has an interior for at least one occupant, at least one vehicle sensor recording occupant impact sensor data describing the movement of at least one of the at least one occupant and/or forces acting on the occupant, and a storage means. In addition, the invention relates to a motor vehicle and an emergency information system.

Motor vehicles in which at least one occupant is travelling in the interior of the vehicle can be the victim of accidents which can also result in personal injury. In the event of such accidents, emergency services are usually alerted and provide initial assistance on site as part of an accident rescue. In a classic, well-known scenario, emergency services or their employees are confronted with the specific situation for the first time after they have reached the scene of the accident and must decide which course of action is most promising based on the picture presented to them. In this case, there is often no information whatsoever about the occupants involved in the accident or their general state of health/history, just as the course of events, particularly with regard to occupants, can at best be gleaned from the impression gained by looking at the scene of the accident. Accordingly, in such a scenario, the time until the emergency services reach the scene of the accident is not useful.

Therefore, there is little or no health information available about at least one of the affected occupants, such as the number of occupants in the vehicle or important patient information such as previous illnesses, medications and allergies.

It has therefore already been proposed in the state of the art that additional information about the occupants can be provided automatically with an emergency call. EN 10 2019 126 208 A1 an emergency system for a vehicle, which has a data collection unit for collecting emergency data, the emergency data comprising at least vehicle data and health data of at least one user of the vehicle, a data storage unit for storing the emergency data and a data access unit via which a communication connection to the data storage unit can be established. This is based on automatic emergency call systems (so-called eCall systems) that automatically or manually make an emergency call after an accident and establish a telephone voice connection to an emergency call center. Emergency vehicles can therefore be alerted quickly. EN 10 2019 126 208 A1 now proposes not only transmitting vehicle data, such as a current image from an interior camera and the status of restraint systems, but also health data of the vehicle users, which should allow conclusions to be drawn about the type and severity of the vehicle users' injuries. The number of vehicle occupants should also be transmitted so that the required number of emergency vehicles can be made available. Health data, which can also be recorded using sensors, can include values on age, gender, height, weight, blood type, body mass index (BMI), allergies, illnesses, implants, pulse, blood pressure, blood sugar, blood transfusion history, transplant history, visual and/or hearing aids, pacemakers, blood alcohol and/or drug influence.

The transmission of such health data or health information of at least one of the at least one occupants of the motor vehicle already represents an improvement, but may not be sufficient in some cases. For example, only little information can be derived from this about the course of the accident, i.e. shortly before the accident (pre-crash phase) and shortly after the accident (post-crash phase). It is also not possible or difficult to classify the accident (front, rear, side impact or rollover) and the resulting or predicted type of injury and/or severity of injury (without the use of artificial intelligence).

EN 10 2018 118 129 A1 relates to an occupant protection system for a motor vehicle, which has an active restraint system, a detection device for detecting a position of a vehicle occupant and a corresponding data processing device. The data processing device is designed to simulate, for at least one accident scenario, based on the detected position of the vehicle occupant, the movements of the vehicle occupant relative to the active restraint system and a triggering of the active restraint system using a predetermined simulation model and to determine a triggering strategy for the active restraint system depending on a result of the simulation for improved protection of the vehicle occupant.

The invention is therefore based on the object of providing an emergency information system that allows better assistance by rescue services.

This object is achieved according to the invention by a method, a motor vehicle and an emergency information system according to the independent patent claims. Advantageous embodiments emerge from the subclaims.

A method of the type mentioned at the outset is characterized according to the invention in that at least one health characteristic of the occupant and the occupant impact sensor data related to at least a predefined retention period are stored in the storage means for generating an emergency information data record, and if an accident signal indicating an accident is present, the emergency information data record is released for retrieval by and/or transmission to a rescue service. The method according to the invention can in particular be a computer-implemented method.

By providing occupant impact sensor data and/or contents of the emergency information dataset derived therefrom, in particular through evaluation, it is possible to expand the knowledge of the emergency services about the course of the accident in relation to the at least one occupant, in particular to understand the course of the accident in relation to the occupant and, if necessary, to deduce the type and/or severity of the injury in advance, for example when approaching the scene of the accident. This is also possible in particular if the at least one occupant is unconscious or unresponsive. The occupant impact sensor data can in particular describe how the occupant moved during the accident in order to be able to determine the specific type and/or severity of the injury, for example depending on the type of acceleration or direction of attack, and/or also which intrusion took place in the interior of the motor vehicle, for example with regard to injuries caused by other vehicle components penetrating the occupant or contact between the occupant and the occupant. It is also possible, for example, to understand which forces acted on the at least one occupant and in which direction of attack.

• - insbesondere mit einer Inertialplattform des Kraftfahrzeugs aufgenommene Kraftfahrzeugbewegungsdaten,
• - mit einer auf den Insassen gerichteten Innenraumsensorik, insbesondere einer Insassenkamera und/oder einem Insassenradarsensor und/oder einem Insassenlasersensor, aufgenommene Insassensensordaten,
• - eine Verformung des Innenraums beschreibende, insbesondere von einer Aufprallsensorik aufgenommene, Verformungsdaten, und
• - auf die Auslösung wenigstens eines Insassenrückhaltesystems, insbesondere eines Airbags und/oder eines Gurtstraffers, bezogene Rückhaltedaten,

gewählt sind.It was recognized that, particularly when combined with health information, an excellent information basis can be provided for the rescue service, which further improves the preparation and implementation of assistance. In other words, the information on the course of the accident relating to at least one occupant and the health information allow the rescue service to be better prepared for the operation, as well as improved accident rescue and specific occupant treatment at the scene of the accident and also later, for example in a hospital. In concrete terms, it can be provided, for example, that the occupant impact sensor data is at least partially taken from the group comprising

• - in particular, vehicle movement data recorded using an inertial platform of the motor vehicle,
• - occupant sensor data recorded with an interior sensor system aimed at the occupant, in particular an occupant camera and/or an occupant radar sensor and/or an occupant laser sensor,
• - deformation data describing a deformation of the interior, in particular recorded by an impact sensor, and
• - restraint data relating to the activation of at least one occupant restraint system, in particular an airbag and/or a belt tensioner,

are selected.

Of course, the occupant impact sensor data can also be pre-evaluated information or evaluation data, in particular derived from raw sensor data and/or signals from vehicle communication.

For example, in order to understand the course of the accident with regard to at least one occupant, occupant impact sensor data from sensors directed at the vehicle interior, in particular the occupant, can be used. Such interior sensor technology can, for example, comprise at least one occupant camera, in particular an occupant observation camera, but additionally or alternatively also further sensors, for example radar sensors, laser sensors and the like. The occupant camera can in particular also be a 3D camera and/or a time-of-flight camera (TOF camera). Such a three-dimensionally measuring camera can also be understood as a depth sensor. Another useful source of information when understanding the course of the accident, in particular with regard to forces acting on the occupant, is the overall movement of the motor vehicle and thus also of the vehicle components delimiting the interior and/or provided in the interior. Modern motor vehicles often comprise an inertial measuring unit, in particular an inertial platform, in order to be able to record data on the vehicle's own movement, in particular on acceleration. The vehicle movement data can particularly advantageously include acceleration data, which are particularly relevant with regard to forces acting on the occupant. In particular, the motor vehicle movement data can therefore include a longitudinal acceleration and/or a lateral acceleration and/or at least an angular acceleration. For example, such vehicle movement data can also be provided indirectly, for example via an ego data control unit of the motor vehicle. Inertial platforms are often also referred to as "inertial measurement units" or IMU.

Other useful occupant impact sensor data are deformation data, which can of course be derived indirectly from other raw sensor data, but in particular can also be provided at least partially by an impact sensor system. An impact sensor system, for example comprising collision sensors or crash sensors, can also form at least part of the basis for generating the accident signal, since such impact sensors are often used to detect a collision and, if necessary, to control and/or precondition other vehicle systems accordingly, for example occupant restraint systems. Corresponding deformation data are therefore used for another useful purpose within the scope of the present invention.

Finally, as an example of occupant impact sensor data, restraint data can also be mentioned that relates to the activation of at least one occupant restraint system, in particular an airbag and/or a belt tensioner. Such occupant restraint systems also act on the occupant before or during the accident and influence their movement and/or the forces acting on the occupant, so that they are also particularly advantageously taken into account when generating the emergency information data set. It should be noted that restraint data can also be used at least partially to generate the accident signal, since, for example, the activation of airbags is often at least taken into account as part of accident detection.

Overall or generally, it can be said that the occupant impact sensor data in their entirety, when evaluated accordingly, preferably allow at least a partial reconstruction of the accident sequence related to the at least one occupant. Different aspects can be covered by different types of occupant impact sensor data, so that, for example, occupant sensor data can also be used to generate occupant models or person models and deformation data and related information can describe impact directions, which also applies to motor vehicle movement data.

Vehicle sensors for recording occupant impact sensor data can therefore include, in particular, interior sensors, impact sensors, an inertial platform (IMU) and/or internal sensors of occupant restraint systems.

The health information can expediently be selected at least partially from the group comprising at least one anamnesis information and at least one medication information. The anamnesis information can in particular include previous illnesses and/or allergies. It is therefore particularly advantageous if the health information describes at least previous illnesses, medications and allergies, as this allows the assistance to be tailored to the specific occupant during preparation and implementation and, with particular advantage, a connection can also be established to the occupant impact sensor data, since some occupants can be more susceptible to certain types of injuries or impacts. However, the health information particularly preferably also includes further occupant information, for example regarding age, gender, height, weight and the like, as is already the case, for example, in EN 10 2019 126 208 A1 was proposed.

It should be noted at this point that it is particularly preferred to record both the health information and the occupant impact sensor data, as far as possible, for all current occupants of the motor vehicle. While this may not always be possible for the health information, it is sensible to implement this at least for the occupant impact sensor data. However, embodiments are also conceivable in which the availability of both the health information and the occupant impact sensor data for all occupants is ensured or at least made possible. In this context, it should be noted that it is particularly advantageous anyway if at least the emergency information data set is also generated to include information about the total number of occupants in the motor vehicle.

It is particularly advantageous for the emergency information data record to be transferred to a server external to the vehicle and made available for retrieval there and/or, in particular if a communication connection to the server cannot be established, made available for retrieval directly from the vehicle, in particular via WLAN, and/or transmitted as part of an emergency call and/or as an emergency call. It is therefore particularly expedient if the emergency information data record is transferred to a server after it has been generated and the accident signal has been received, from where it can be retrieved. The availability of the emergency information data set is then independent of the further fate of the motor vehicle and its systems, and a greater range is also achieved by the server external to the vehicle. The server can be designed as a cloud or part of a cloud, for example. It also allows the emergency information data set to be called up multiple times, for example in several rescue service vehicles, for preparation from a hospital and the like. In addition, when stored externally on a server that is ideally used by a large number of motor vehicles, other functions based on emergency information data sets can be implemented, such as storage, statistical analysis and the like, particularly in the sense of an accident database. A server can also provide the computing power to evaluate the emergency information data set and, if necessary, to enrich it with evaluation results; in addition, an evaluation of emergency information data sets that are received and provided on the server can also be evaluated or used with a view to improving the function of the motor vehicle.

If it is determined that no communication connection to the server can be established, for example due to damage, a direct retrieval can be enabled on the vehicle side, particularly via WLAN, so that the emergency services can access it in this way. In principle, it is of course also conceivable to attempt a transfer to the server and also to enable local retrieval directly from the vehicle's storage device in each case.

Additionally or alternatively, it can be provided that the emergency information data record is transmitted as part of an emergency call or as an emergency call. For example, if the emergency information system is implemented in conjunction with or as part of an emergency call system, it is expedient to also send the information in the emergency information data record with the emergency call and thus make it available. The emergency information data record can also be expanded to fully represent the emergency call, and thus be sent as an emergency call. The emergency information data record can then also contain further emergency information, for example a geodetic position of the motor vehicle, a number of occupants, motor vehicle data and the like.

In an expedient further development, it can be provided that the storage means, in particular together with at least part of an associated control device that carries out the method, is installed in a crash-safe location in the motor vehicle. From a mechanical point of view, the storage means, in particular together with at least part of a control device of the motor vehicle that carries out the method, can be installed in such a way that damage from an accident, in particular a crash, is as unlikely as possible, for example centrally in the motor vehicle in the area of an underbody. For example, techniques such as those known from so-called black boxes or tachographs can be used here. The storage means and at least part of the control device that carries out the method can also be assigned an emergency battery, in particular in a corresponding structural unit, in order to be able to generate, provide and/or send the emergency information data set in the event of a failure of the electrical power supply to the on-board network.

In particular when transmitted as and/or as part of an emergency call, but also in other cases, current geodetic position information of the motor vehicle, in particular a GPS position, can be added to the emergency information data record when it is generated. Such position information can be provided, for example, by a position determination device of the motor vehicle, for example a position sensor.

• - von dem wenigstens einen und/oder wenigstens einem weiteren des wenigstens einen Insassen mittels eines kraftfahrzeuginternen und/oder kraftfahrzeugexternen Eingabemittels eingegeben und/oder
• - von einer externen Informationsquelle, insbesondere unter Nutzung einer auf einem mit dem Kraftfahrzeug verbundenen Mobilgerät des wenigstens einen Insassen genutzten Gesundheitsapplikation, automatisch abgerufen werden.

The health information can be at least partially

• - entered by the at least one and/or at least one other of the at least one occupant by means of an input means inside and/or outside the vehicle and/or
• - be automatically retrieved from an external information source, in particular using a health application used on a mobile device connected to the motor vehicle by at least one occupant.

In other words, different sources can be used for the health information. On the one hand, input by at least one passenger, for example the driver, is possible, whereby both vehicle-internal and vehicle-external input means can be used. An internal vehicle input means can be implemented, for example, by a human-machine interface, for example a touchscreen on the instrument panel and/or a voice input system. In a particularly expedient embodiment, however, a control system assigned to the vehicle and based on An application running on a mobile device that can be connected to the motor vehicle, for example a motor vehicle app, can be used to query health information. Such motor vehicle apps have already been proposed in principle in the prior art and allow, for example, parameters of the motor vehicle to be set and/or the motor vehicle to be remotely controlled, for example with regard to pre-climate control. They can easily be expanded to include the option of providing health information.

However, it is also expedient and preferred according to the invention to provide for at least partially automatic retrieval of at least part of the health information from an external information source, for example a mobile device connected to the motor vehicle. For example, an embodiment of the present invention can provide for synchronization with a health application running on a mobile device connected to the motor vehicle, for example a health application from a health insurance company, to retrieve the health information. For example, access to an electronic patient file can be made possible using the health application. In particular, such synchronization with health applications is also possible for several occupants of the motor vehicle, for example if, as already proposed in the prior art, mobile devices of different occupants can be detected, addressed communicatively and also assigned to occupants, for example by locating or evaluating transmission and/or reception strengths. It is then possible to retrieve the health information of the respective occupant from a mobile device assigned to the respective occupant, in particular by synchronization with a health application.

In an advantageous development of the present invention, the accident signal can be generated when at least one occupant restraint system, in particular at least one airbag, is triggered and/or by evaluating sensor data from at least one impact sensor, and/or the emergency services can be automatically alerted with the accident signal by means of an emergency call. Options for detecting an accident, for example a crash, are already known in principle in the prior art. For example, an accident can be detected based on the triggering of airbags, whereby data from other safety systems can also be used or included. For example, as already mentioned, it is known to equip the motor vehicle with an accident sensor system, in particular impact sensors, in order to be able to detect a collision or an accident in general. Once the accident signal is provided, not only is the emergency information data set generated, but an emergency call can also be made automatically, whereby, as already mentioned, the emergency information data set can certainly be used as part of this emergency call or can even form the emergency call itself. However, within the scope of the present invention, it is also possible to use a manual emergency call system or emergency system, for example a dedicated emergency call button inside the motor vehicle. In this case, at least in some of the conceivable cases, the generation of the accident signal can also be linked to the activation of such a manual emergency call system or emergency system.

Preferably, a ring buffer dimensioned according to the length of the retention period can be used to store the occupant impact sensor data. The storage means can therefore comprise a ring buffer in which the occupant impact sensor data are continuously stored during operation of the motor vehicle, so that they are always available for the retention period and no complicated control is necessary.

The length of the retention period can be, for example, 10 to 20 seconds. A length of 15 seconds is particularly preferred. Such a period is usually sufficient to be able to record the course of an accident as completely as possible. It can be particularly advantageous that after the accident has occurred, in particular after the accident signal has been received, occupant impact sensor data for the emergency information data set is recorded and stored for a predefined partial time period, in particular 4 to 6 seconds, of the retention period. The idea behind this is that once the accident has been recognized, it does not necessarily have to be over, in particular with regard to the impact on the occupant, but relevant incidents can still exist afterwards, for example until the motor vehicle comes to a complete standstill and/or due to the impact of occupant restraint systems. The partial time period can in particular be 5 seconds. An advantageous combination results when the retention period is 15 seconds long, 5 seconds of which correspond to the partial time period. Then 10 seconds before an accident signal, for example a crash, and 5 seconds after the accident signal, for example the crash, can be recorded.

When creating the emergency information dataset, it is not just a matter of compiling it, but of An evaluation process may already be taking place, for example with a view to reducing the amount of data, in particular with regard to data portions that are not relevant to the accident or the impact on the occupants. In other words, the emergency information data set may at least partially comprise evaluation information determined by evaluating the occupant impact sensor data and/or the health information. This may include simpler evaluation processes as well as more complex, modeling evaluation processes. Simple evaluation processes include, for example, reducing the sensor data to certain portions, for example portions indicating an occupant and/or portions relating to the impact on the occupant.

In a specific, particularly preferred embodiment of the invention, it can be provided that, in order to determine at least part of the occupant impact sensor data and/or the emergency information data set, in particular after the accident signal is present for generating the emergency information data set, raw sensor data relating to the interior of the motor vehicle from at least one of the at least one vehicle sensor, in particular the interior sensor of the interior sensor system, are evaluated to form a static human model of the at least one occupant and/or a static interior model. Such an evaluation can, for example, be carried out for each point in time covered by the retention period, so that a temporal sequence of static human models and/or interior models that can depict the course of the accident can result in particular as evaluation information. However, a time-related evaluation is also conceivable, for example in the case of penetrating vehicle components. While it would in principle be conceivable to carry out this determination of models continuously, in particular in real time, and to store the resulting models in the storage means, in particular the ring buffer, it may be preferable to only carry out such evaluation processes for the retention period when the accident signal is also present in order to reduce complicated calculation processes and thus the computing power required. Corresponding evaluation techniques have already been proposed in the prior art, can use artificial intelligence, for example, and can also be used accordingly in the context of the present invention. For example, a camera and/or other interior sensors can provide an image of the interior of the motor vehicle. Raw sensor data from various interior sensors, in particular interior sensors based on different measurement principles, can be fused with particular advantage in order to be able to generate the static human model and the static interior model.

In this context, it is particularly preferred if the human model and the interior model are evaluated for generating and/or evaluating the emergency information data set to determine contact points between the human model and the interior model and/or, in particular together with acceleration data of the motor vehicle, forces acting on the human model. On the basis of the at least one human model and the interior model, contact points between the at least one human model and the interior model can therefore particularly preferably be calculated, in particular when acceleration data including the acting forces, including their angle of attack, are available. This provides important and essential information when assessing the course of the accident and when estimating the severity and/or type of injury, which may be carried out by the emergency services. Health information and possibly other occupant information can also be included at this point in time, in particular when generating static human models.

Independently of the models mentioned, but preferably in combination with them, an advantageous development of the method can provide that the acceleration data and/or sensor data from an impact sensor system are evaluated to determine classification information describing the type of accident, in particular an impact direction. From the acceleration effect, in particular in the horizontal plane, but also from sensor data from the impact sensor system, an impact direction (attack direction) can be determined in the event of a collision, which can already provide clear indications of possible injuries to occupants. In general, the movement of the motor vehicle as a whole can also provide a good indication of the type of accident. In particular, it can be provided that the acceleration data and/or the sensor data from the impact sensor system are evaluated to determine an accident type as classification information, with the accident type being selected from the group comprising front impact, rear impact, side impact and/or rollover.

In a frontal collision, the occupants usually move further forward due to inertia, which can lead to contact with interior elements located in front of the respective occupants, such as the steering wheel, dashboard or, in the case of occupants in a rear seat, with a seat in front of them. In addition, injuries due to the seat belt. Typical injuries include, for example, traumatic brain injuries, vertebral fractures and injuries to internal organs/vascular ruptures.

In a rear-end collision, passengers usually move backwards first and can then be moved further forwards, for example in a rear-end collision. This can often result in whiplash or other injuries to the cervical spine.

In the case of side impacts, occupants move towards the side of the vehicle. This can result in injuries to the pelvis and/or shoulder or to internal organs, for example due to the penetration of interior components. Finally, in the case of a rollover, possible injuries such as head injuries or, in general, multiple injuries due to various points of contact with the interior can be assumed, particularly when the human model and the interior model are also considered.

It should be noted at this point that at least some of the raw sensor data, in particular occupant impact sensor data from a camera, can of course continue to be part of the emergency information data set, in particular in addition to the evaluation results or evaluation information. This preferably also applies to acceleration data.

• - Sensorrohdaten wenigstens einer Insassenkamera, insbesondere also ein Kamerabild des Innenraums mit den Insassen, insbesondere Fahrer und Mitfahrer,
• - Auswertungsinformationen, die durch Auswertung von Rohsensordaten, beispielsweise aus Rohsensordaten der wenigstens einen Insassenkamera und weiterer Innenraumsensoren, ermittelt wurden, umfassend insbesondere wenigstens ein ermitteltes Menschmodell und/oder wenigstens ein ermitteltes Innenraummodell und/oder berechnete Kontaktpunkte zwischen dem wenigstens einen Menschmodell und dem Innenraummodell, bevorzugt einwirkender Kräfte mit Angriffswinkel auf bevorzugt jeden Insassen im Innenraum, ggf. auch die Klassifizierungsinformation,
• - Beschleunigungsdaten, insbesondere umfassend gemessene Beschleunigungen, Winkelbeschleunigungen und/oder Angriffswinkel des Unfalls bezogen auf das Kraftfahrzeug, und
• - die Gesundheitsinformationen.

For example, in a specific embodiment, the emergency information data set may contain at least:

• - raw sensor data from at least one occupant camera, in particular a camera image of the interior with the occupants, in particular the driver and passengers,
• - evaluation information determined by evaluating raw sensor data, for example from raw sensor data of the at least one occupant camera and other interior sensors, comprising in particular at least one determined human model and/or at least one determined interior model and/or calculated contact points between the at least one human model and the interior model, preferably forces acting with angles of attack on preferably each occupant in the interior, possibly also the classification information,
• - Acceleration data, in particular comprehensive measured accelerations, angular accelerations and/or angle of attack of the accident related to the motor vehicle, and
• - the health information.

In a preferred embodiment of the method according to the invention, as part of the emergency information data set, prediction information concerning possible injuries to at least one occupant, in particular injury severity prediction information and/or injury type prediction information, can be determined by means of a trained prediction function which uses at least part of the occupant impact sensor data as input data.

Artificial intelligence can therefore also be used to assess, particularly pre-assess, the course of the accident in order to estimate any injuries that may have occurred. In general, a trained function represents cognitive functions that people associate with other human brains. Through training based on training data, so-called machine learning, the trained function is able to adapt to new circumstances and detect and extrapolate patterns. Parameters of a trained function can therefore be adjusted through training. Known learning techniques include supervised learning, semi-supervised learning, unsupervised learning, reinforcement learning and/or active learning, although representation learning (“feature learning”) can also be used. During training, the parameters of the trained function are adjusted, in particular iteratively, through several training steps.

The trained detection function can advantageously be a neural network, in particular a deep neural network and/or a convolutional neural network (CNN). However, the use of support vector machines (SNM), decision trees and/or Bayesian networks is also conceivable. The trained prediction function can be based on K-means clustering, Q-learning, genetic algorithms and/or assignment rules, for example. The use of so-called "adversarial networks", for example a GAN (Generative Adversarial Network), is also conceivable.

The use of artificial intelligence, i.e. the trained prediction function, to predict the type and severity of injury allows a prediction of what can be expected during the preparation phase or at least at the scene of the accident. This enables the emergency services to prepare more precisely for the operation, for example to assess whether the accident is serious or a relatively minor one and/or which doctors may be needed.

While it is fundamentally conceivable to use the trained prediction function within the motor vehicle, so that the prediction information can already become part of the emergency information data set before it is provided or released, it can also be expedient to use the trained prediction function outside the vehicle, for example on the server already mentioned. Then an up-to-date version of the trained prediction function can always be kept for all motor vehicles and greater computing power is available, especially if the server forms a cloud and/or part of a cloud. It is particularly advantageous to train the prediction function on such a server, especially using actually recorded emergency information data sets.

Thus, an expedient development of the present invention provides that each emergency information data record is stored externally to the motor vehicle, in particular on the server, and at least one result information relating to injuries to at least one occupant is assigned to the emergency information data record at a later point in time in order to form a training data record for training the trained prediction function. For example, it is therefore conceivable to store the emergency information data record at least until a diagnosis is made, for example by the hospital, which allows subsequent labeling, in particular according to the type and severity of the injury. Such an assignment of result information, which can describe the injuries that actually occurred and their severity, can also be carried out at the scene of the accident by the emergency services or subsequently in the hospital or by a treating doctor. Once the result information has been assigned to the stored emergency information data record, training data can be created from it in order to train the trained prediction function. For example, a software application that can communicate with the server can be used for labeling. If the trained prediction function is already in use, a comparison can be made with its prediction information and taken into account during training. Using the procedure described, it is therefore not only possible to fundamentally train the trained prediction information, but also to continuously improve it.

Typical input data for such a trained prediction function can be formed, for example, by camera images, acceleration data and/or other sensor data from interior sensors. Evaluation information can be used as input data with particular advantage, for example the models described (human model, interior model) and other evaluation information determined with these, for example impact forces. Parts of the health information can also form input data for the trained prediction function.

In a further development, it can also be provided that the trained prediction function, in particular based on the prediction information and/or when using the health information as input data, also issues a triage recommendation in order to further support the emergency services. Such a triage recommendation can, for example, contain a suggestion as to which occupants are to be rescued and cared for in which order.

In particular, when determining classification information that describes the type of accident, it is also possible to determine predictive information on the type and/or severity of injury without the use of artificial intelligence, for example on the basis of the assignment discussed above.

In an advantageous development of the present invention, it can be provided that the retrieved emergency information data set, in particular the occupant impact sensor data, is evaluated on a display device of the rescue service for an optical visualization of the impact on the at least one occupant, in particular for a representation of the processes in the interior relating to the at least one occupant. Such an embodiment is particularly useful when determining further evaluation information, in particular, as described, the at least one human model and the interior model. For example, in the optical visualization, the human models of the occupants with contact points in the vehicle interior (based on the interior model) including the impact forces can be displayed, for example by using a false color display. A preparation of the emergency information data set in a representation can in particular compile several essential pieces of information, for example as a user interface in which a camera image/camera video of the interior is displayed parallel to a visualization of evaluation information, in particular human models in the interior model, and health information. The preparation of the emergency information data set can, for example, be carried out via a software application, in particular a notification app. For example, the emergency services can initially receive a notification with essential data, in particular the geodetic position information of the vehicle, which is stored in a map. In one embodiment, it is conceivable to display the accident as an interactive event representation in a map display, whereby after interaction with the event representation, the optically visualized representation or the user interface is displayed. The emergency services can then be dispatched to the rescue.

In addition to the method, the present invention also relates to a motor vehicle, having a control device designed to carry out the method according to the invention and the storage means. All statements relating to the method according to the invention can be transferred analogously to the motor vehicle according to the invention, with which the advantages already mentioned can also be obtained.

In particular, as already described, a structural unit comprising at least parts of the control device and the storage means can be installed in the motor vehicle in a manner as safe as possible from accidents, for example in the underbody area in the middle underneath or in the interior.

Finally, the invention also relates to an emergency information system comprising a motor vehicle according to the invention and a server external to the motor vehicle. Here too, the statements on the method and the motor vehicle apply accordingly. In an expedient embodiment, the emergency information system can also comprise at least one computing device of at least one rescue service, on which, for example, the software application described can be executed in order to enable optical visualization of contents of the emergency information data set. The at least one computing device of the rescue service can also comprise mobile devices of the rescue service in order to enable preparation for the accident as early as the approach to the scene of the accident. The server can, as already described, be part of a cloud or form a cloud itself. It can have a storage device for storing emergency information data sets. In particular, a training unit can also be provided on the server side in order to be able to train the prediction function.

• 1 einen Ablaufplan eines Ausführungsbeispiels des erfindungsgemäßen Verfahrens für Schritte, die innerhalb eines Kraftfahrzeugs durchgeführt werden,
• 2 einen Ablaufplan für außerhalb des Kraftfahrzeugs durchführbare Schritte,
• 3 ein erfindungsgemäßes Kraftfahrzeug, und
• 4 ein erfindungsgemäßes Notfallinformationssystem.

Further advantages and details of the present invention emerge from the exemplary embodiments described below and from the drawings.

• 1 a flow chart of an embodiment of the method according to the invention for steps that are carried out within a motor vehicle,
• 2 a schedule of steps that can be carried out outside the vehicle,
• 3 a motor vehicle according to the invention, and
• 4 an emergency information system according to the invention.

1 shows a flow chart of an embodiment of the method according to the invention, insofar as it is carried out in a motor vehicle. The motor vehicle has an interior in which occupants, for example including a driver, can be present. In order to prepare the generation of an emergency information data record in the event of an accident, health information of at least one of the occupants, in particular all occupants, is obtained in a step S1 and stored in a storage means of the motor vehicle. In this case, there are various possibilities for obtaining the health information, particularly preferably together with other occupant information, which may also be used alternatively or in addition to one another within the scope of the present invention.

On the one hand, it is conceivable that before starting a journey, for the driver, for example, also permanently for a plurality of journeys, health information of at least one of the at least one occupants is received by means of an input device external to the motor vehicle and/or internal to the motor vehicle. An input device internal to the motor vehicle can, for example, be part of a human-machine interface of the motor vehicle, for example a touchscreen provided in the motor vehicle. Preferably, however, an input device external to the motor vehicle can be used, for example a mobile device of at least one of the at least one occupant, for example the driver. For this purpose, an application assigned to the motor vehicle, for example a motor vehicle app, can be run on the mobile device, with which other information can also be transmitted from and to the motor vehicle and/or functions of the motor vehicle can be controlled, for example remote control of the air conditioning of the interior.

In another possibility, the health information, possibly together with other passenger information, is automatically retrieved from an information source external to the vehicle, for example at least one mobile device. A health application (health app) can be run on the corresponding mobile device, which is then assigned to the at least one passenger, which, for example, provides access to an electronic patient file. of the occupant and/or health information of the occupant provided by a health insurance company. In this sense, the health information in the storage means of the motor vehicle is then synchronized with the health information of the health application on the mobile device. It is particularly expedient in this context if mobile devices that can communicate with the motor vehicle can be detected within the interior and assigned to corresponding occupants, the number of which can also be determined in another way, so that in this way health information of a plurality of occupants, in particular all occupants, can be automatically determined. If, for example, a number of occupants determined in another way, which can also be stored in the storage means, deviates from the number of occupants for whom health information could be retrieved, at least one of the at least one occupant, in particular the driver, can be requested at the input means inside or outside the vehicle to provide the missing health information manually or in another way.

During a journey or an operating phase in which the occupants do not change, their health information, along with the number of occupants and, if applicable, other occupant information, remains permanently stored in the storage medium.

In the present case, health information includes at least anamnesis information and at least medication information for the passenger, thus describing in particular previous illnesses, allergies and medication taken by the passenger. For example, health information can describe whether the passenger suffers from high or low blood pressure, wears implants such as a pacemaker, regularly takes medication for a chronic illness and the like. All of this can be important information for emergency services in the event of an accident.

In the present case, the storage means also has a ring buffer, which is used in a step S2 to continuously store occupant impact sensor data, at least for a predefined retention period. The length of the retention period is selected so that the processes that occur in an accident and that relate to the occupants can be reproduced as completely as possible. For example, the length of the retention period can be between 10 and 20 seconds, in particular 15 seconds.

The occupant impact sensor data, which are stored in the ring buffer for at least the retention period, are suitable for reconstructing impacts on occupants that occur during an accident. This can be raw sensor data, pre-evaluated sensor data or information available elsewhere in the motor vehicle, for example bus signals. In the present case, the occupant impact sensor data includes motor vehicle movement data, in particular acceleration data, which can be recorded with an inertial measurement unit (IMU), i.e. an inertial platform, of the motor vehicle. The motor vehicle movement data, in particular acceleration data, can be queried directly from the inertial platform, but can also be provided by a control unit of the motor vehicle that monitors the motor vehicle's own movement and possibly evaluates it in advance.

The occupant impact sensor data also includes occupant sensor data that is recorded using an interior sensor system aimed at the occupant(s). In the present case, the interior sensor system includes, for example, at least one occupant camera and at least one occupant radar sensor and/or at least one occupant laser sensor in order to also obtain three-dimensional information. The occupant camera can also be designed as a 3D camera, in particular a TOF camera. For example, at least one of the at least one occupant camera can be a driver observation camera; however, at least one general interior camera that captures the entire interior, for example using a wide-angle lens, is also conceivable, and which of course also captures the occupants. Here, too, a preliminary evaluation can already be present, in particular with regard to the occupant radar sensor and/or the occupant laser sensor, for example by storing recorded surface profiles of the interior in the ring buffer instead of the raw sensor data. Occupant impact sensor data can also generally already be transferred to a common coordinate system, in particular a motor vehicle coordinate system.

The occupant impact sensor data can also include deformation data describing a deformation of the interior, in particular deformation data recorded by an impact sensor system. An impact sensor system can, for example, include collision sensors on the body and/or other parts of the vehicle body shell, which record the forces acting on it and can therefore also indicate whether and when there is a deformation of the interior or the penetration of other vehicle components into the The sensor data from such an impact sensor system therefore also represent useful information, even if the corresponding information can possibly be derived at least partially from data from the interior sensor system.

The latter also applies to restraint data relating to the activation of at least one occupant restraint system that is optionally included in the occupant impact sensor data. For example, the activation of an airbag and/or a belt tensioner can be logged in the ring buffer.

It should be noted that different levels of evaluation are basically possible for determining the occupant impact sensor data to be stored in the ring buffer, in particular in order to keep the amount of data to be stored as low as possible without losing any information regarding the impact forces acting on the at least one occupant. For example, irrelevant information can be filtered out and/or the occupant impact sensor data can be converted into a data format that reduces the amount of data.

In a step S3, it is checked whether an accident signal indicating an accident is present. If this is not the case, the ring buffer is filled for the retention period in accordance with step S2. If this is the case, an emergency information data set is generated from at least the health information and the occupant impact sensor data in step S4 and released for retrieval or transmitted to an emergency service. The accident signal can be generated in different ways, possibly also in combination. For example, the accident signal can be generated when at least one occupant restraint system, in particular at least one airbag, is triggered. In addition or alternatively, sensor data from the impact sensor system can be evaluated. In the present case, automatic generation of the accident signal is preferred in any case, although manual generation, for example using an emergency button on the motor vehicle, is also conceivable in principle.

In step S4, the emergency information data record is then generated. To do this, the recording of the occupant impact sensor data, i.e. their storage in the storage medium, is continued for a predefined partial time period, in this case 5 seconds, of the retention period. This takes into account the fact that the accident and thus also the impact on the occupant is not necessarily already over when the accident signal is present. The occupant impact sensor data describing the impact on at least one occupant is therefore completed.

To generate the emergency information data set, the occupant impact sensor data, which are available in the ring buffer after data recording has been completed, are then evaluated in order to determine evaluation information for the emergency information data set. For this purpose, occupant impact sensor data relating to the interior of the motor vehicle, in particular raw sensor data, are evaluated in order to determine statistical human models for each of the occupants at any time that is included in the specified period and is therefore stored in the ring buffer, just as a statistical interior model is determined for each time point. The interaction of different measurement principles can pay off in particular if both image data, for example from an occupant camera, and 3D data, for example from an occupant laser sensor and/or an occupant radar sensor or due to the use of a 3D camera, are available. Corresponding occupant impact sensor data are fused and provide human models for the occupants. A statistical model of the interior can also be generated as part of this data fusion. Changes over the different points in time show the effects of the accident. In particular, a further evaluation is carried out to calculate contact points between at least one human model and the interior model, from which acting forces and angles of attack can then be identified. For the latter step, the acceleration data can also be included. Deformation data can also be useful with regard to contact points. At least the result (contact points and acting forces including angle of attack) is added to the emergency information data set, which expediently also contains at least some of the occupant impact sensor data, for example at least camera images of the interior with the occupants. Measured accelerations, angular accelerations and angles of attack of the accident in relation to the motor vehicle, in particular vehicle movement data, in particular acceleration data, and deformation data, can also be added to the emergency information data set. This also applies in any case to the health information, which is also stored in the storage medium. Finally, geodetic position information for the motor vehicle, in this case a GPS position, is also added to the emergency information data set.

In addition to the described evaluation based on the model or alternatively to this It can also be provided that by evaluating the acceleration data and/or the sensor data from the impact sensor system, classification information describing the type of accident is determined and added to the emergency information data set. The type of accident, for example frontal impact, rear impact, side impact, rollover or the like, also allows conclusions to be drawn about possible injuries.

In a step S5, it is then checked whether a server to which the emergency information data record is to be transmitted for retrieval by a rescue service is reachable, and therefore whether a communication connection can be established there. This may not be the case, for example, if the motor vehicle is too badly damaged and/or there are shields, e.g. in a tunnel. If the server is reachable, the emergency information data record is forwarded to it; see step S6. The emergency information data record simultaneously forms an emergency call, which means that when the emergency information data record is received on the server, the rescue service is automatically alerted and finds all the information it needs in the emergency information data record. It should be noted, however, that in an optional step S8, a direct transmission to at least one computer device of the rescue service is also conceivable.

If the server cannot be reached according to step S5, the emergency information data record is made available for retrieval directly from the vehicle, in this case via WLAN, in a step S7. This means that it can be read directly from the vehicle by the emergency services.

It should also be noted at this point that the storage means and at least part of a control device carrying out the method according to the invention within the motor vehicle, in particular as a structural unit, are mechanically particularly protected in order to withstand the effects of the accident as robustly as possible.

2 shows, as a flow chart, further steps which are carried out by the server or a computer device of the rescue service, which can also form part of a corresponding emergency information system according to the invention.

In a step S9, the emergency information data set is expanded again by further evaluation by the server, in particular in the cloud. In the present exemplary embodiment, a trained prediction function is applied to at least part of the contents of the emergency information data set in order to determine prediction information. After the trained prediction function predicts an injury severity and an injury type, it provides injury severity prediction information and injury type prediction information as output data. At least part of the occupant impact data is used as input data, particularly expediently the contact points and the impact forces as they were already determined in the motor vehicle.

It should be noted at this point that it is of course also conceivable in principle to provide the trained prediction function in the motor vehicle in exemplary embodiments, but this is less preferred because the most recent version can always be available on the server and, in addition, training can also be carried out, if necessary continued, as will be explained in more detail below.

The prediction function can also use health information as input data if this can influence the prediction. It should also be noted that a triage recommendation can also be generated as output data, i.e. it can be specified which passengers/injured people should be cared for first.

The emergency information data record expanded in step S9 is then made available for retrieval by the rescue service in a step S10 and retrieved in a step S11, for example after a notification of the emergency call. In step S12, a representation can be generated on a computer device of the rescue service, which, among other things, also visually visualizes the processes in the interior that affect the at least one occupant. For example, a user interface can be displayed on which the accident site is initially shown on a map based on the geodetic position information, whereby after interaction with the interactive event representation on the map in the user interface, an informative representation is displayed which, in addition to information on the number of occupants and the health information, can display a camera image of the interior of the motor vehicle, in particular as a video during the accident, a graphic representation of the at least one human model, the interior model and the contact points including the impact forces including the angle of attack. In such a graphic representation of the impact forces and contact points, false colors can be used, for example, for affected areas of the occupant or the strength of impact forces.

In a step S13, the emergency information data record is at least partially stored permanently on the server. This can serve a variety of purposes, for example expanding an accident database, creating accident statistics and the like. In the present case, however, the emergency information data records are also used for training the prediction function, possibly continuing to do so. To this end, in a step S14, the emergency information data record is labeled at a later point in time by adding result information that describes at least the severity and type of injury of at least one occupant. Training output data (ground truths) are then available that allow the creation of a corresponding training data record for the prediction function. The corresponding training can be carried out using this training data record in a step S15.

The result information can, for example, come from an initial diagnosis at the scene of the accident itself, i.e. from the emergency services, via a suitable application, and/or can also be provided, again via a suitable application, by a hospital and/or a treating physician.

3 shows a schematic diagram of a motor vehicle 1 according to the invention, which can also be part of an emergency information system according to the invention. The motor vehicle 1 has an interior 2 in which occupants can stay. These occupants can, for example, carry mobile devices 3 that do not belong to the motor vehicle, such as cell phones. By means of a suitable communication device 4 of the motor vehicle 1, which can also establish communication with the server external to the motor vehicle, a communication connection to the mobile devices 3 can also be established, so that, for example, data relating to a motor vehicle application 5 running on a mobile device 3 can be exchanged. At least some of the mobile devices 3 also have a health application 6, which can, for example, access an electronic patient file and/or general health data from a health insurance company. By means of the communication device 4, this health information can be retrieved from the respective health application 6.

To monitor the interior 2 and the occupants, the motor vehicle 1 has an interior sensor system 7, which can, for example, have at least one occupant camera 8, at least one occupant radar sensor 9 and/or at least one occupant laser sensor 10. Their raw sensor data can already represent a first basis of the occupant impact sensor data, as described.

Also accessible from the interior 2 is an input device 11 inside the vehicle, here as part of a human-machine interface 12, via which, for example, health information can be entered manually. In principle, this is also possible using the mobile device 3 as an input device outside the vehicle, in particular using the motor vehicle application 5.

Other systems that can form information sources for the emergency information data set and the accident signal include an impact sensor system 13, an inertial platform 14 (inertial measuring unit, IMU) and occupant restraint systems 15, for example airbags 16 and belt tensioners 17. Current geodetic position information of the motor vehicle 1 can be determined via a GPS sensor 18.

In order to implement the process according to the invention at least in the proportions of 1 In order to be able to carry out the emergency call, the motor vehicle 1 has a control device 19 and a storage means 20, which also includes a ring buffer 21. The storage means 20 and the control device 19 are accommodated as a structural unit 22 in a mechanically stable housing and are installed centrally in the motor vehicle 1, for example in the underbody area in the interior, in order to be as safe as possible from damage in the event of an accident. The structural unit 22 can also contain an emergency battery and/or a WLAN communication device for retrieving the emergency information data record if a connection to the server cannot be established via the communication device 4.

As already mentioned, the control device 19 is designed to carry out the 1 shown steps of the method according to the invention.

4 shows an emergency information system 23 according to the invention, to which several motor vehicles 1 belong, all of which can communicate via communication devices 24 with the server 25, which is implemented here as a cloud by way of example. On the part of the server 25, the steps S9, S10, S13, S14 and S15 can be carried out according to 2 For this purpose, the server 25 can have a storage device 26 in which the current version of the trained prediction function as well as emergency information data records and possibly associated result information can be stored, as well as a prediction and training unit 27 for these steps.

Computer devices 28 of various rescue services, each of which can have a corresponding application 29, for example for retrieving, for visual visualization and also for entering result information, can also communicate according to the arrows 30. The communication connections 24, 30 can be established, for example, via the Internet and/or a mobile network. This also applies to an optional communication connection 31 to a computer device 32 of a hospital or doctor, which in turn can have an application 33 for forwarding result information. The application 33 can also correspond to the application 29 if a doctor or hospital staff also wants to take a closer look at the course of the accident.

In particular, the computing devices 28, 32 can be at least partially mobile devices, for example smartphones and/or tablets, so that data can be exchanged and the data viewed on the move, for example while driving to the accident site or at the accident site.

Finally, it should be noted that the application 29 also configures corresponding mobile devices to retrieve the emergency information data set directly from the motor vehicle 1 via WLAN, in particular when the server 25 is not accessible.

QUOTES INCLUDED IN THE DESCRIPTION

This list of documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions.

Cited patent literature

• DE 102019126208 A1 [0004, 0018]
• DE 102018118129 A1 [0006]

Claims (15)

Method for operating an emergency information system (23) comprising at least one motor vehicle (1), the motor vehicle (1) having an interior (2) for at least one occupant, at least one vehicle sensor recording occupant impact sensor data describing the movement of at least one of the at least one occupant and/or forces acting on the occupant, and a storage means (20), characterized in that in the storage means (20) for generating an emergency information data set - health information describing at least one health characteristic of the occupant and - the occupant impact sensor data related to at least a predefined retention period are stored and, if an accident signal indicating an accident is present, the emergency information data set is released for retrieval by and/or transmission to an emergency service. Procedure according to Claim 1 , characterized in that the emergency information data record is transmitted for retrieval to a server (25) external to the motor vehicle and is made available there for retrieval and/or, in particular if a communication connection to the server (25) cannot be established, is made available for retrieval directly from the motor vehicle (1), in particular via WLAN, and/or is transmitted as part of an emergency call or as an emergency call. Procedure according to Claim 1 or 2 , characterized in that, in particular before or at the start of the journey, the health information is at least partially entered by the at least one and/or at least one further of the at least one occupant by means of an input means (11) inside and/or outside the motor vehicle. Method according to one of the preceding claims, characterized in that, in particular before or at the start of the journey, the health information is at least partially retrieved automatically from an external information source, in particular using a health application (6) used on a mobile device (3) of the at least one occupant connected to the motor vehicle (1). Method according to one of the preceding claims, characterized in that the accident signal is generated when at least one occupant restraint system (15), in particular at least one airbag (16), is triggered and/or by evaluating sensor data from at least one impact sensor (13) and/or the emergency services are automatically alerted using the accident signal as an emergency call. Method according to one of the preceding claims, characterized in that a ring buffer (21) dimensioned according to the length of the retention period is used to store the occupant impact sensor data and/or the retention period is 10 to 20 seconds long and/or after the occurrence of the accident, in particular the presence of the accident signal, occupant impact sensor data for the emergency information data set are recorded and stored for a predefined partial time period, in particular 4 to 6 seconds, of the retention period. Method according to one of the preceding claims, characterized in that the emergency information data set at least partially comprises evaluation information determined by evaluating the occupant impact sensor data and/or health information. Method according to one of the preceding claims, characterized in that in order to determine at least a part of the occupant impact sensor data and/or the emergency information data set, in particular after the presence of the accident signal, raw sensor data of at least one of the at least one vehicle sensor relating to the interior (2) of the motor vehicle (1) are evaluated to form at least one static human model of the at least one occupant and/or a static interior model. Procedure according to Claim 8 , characterized in that the human model and the static interior model are evaluated for generating and/or evaluating the emergency information data set for determining contact points between the human model and the interior model and/or, in particular together with acceleration data of the motor vehicle (1), forces acting on the human model. Method according to one of the preceding claims, characterized in that the occupant impact sensor data are at least partially selected from the group comprising - motor vehicle movement data recorded in particular with an inertial platform (14) of the motor vehicle (1), - occupant sensor data recorded with an interior sensor system (7) directed at the occupant, in particular an occupant camera (8) and/or an occupant radar sensor (9) and/or an occupant laser sensor (10), and - deformation of the interior (2) describing, in particular by an impact sensor system (13) recorded deformation data, - restraint data relating to the activation of at least one occupant restraint system (15), in particular an airbag (16) and/or a belt tensioner (17); and the health information is at least partially selected from the group comprising - at least one anamnesis information and - at least one medication information. Method according to one of the preceding claims, characterized in that, as part of the emergency information data set, a possible injury to the at least one occupant is determined by means of a trained prediction function which uses at least part of the occupant impact sensor data as input data. Procedure according to Claim 11 , characterized in that each emergency information data record is kept external to the motor vehicle, in particular on the server (25), and at least one result information concerning injuries of the at least one occupant is assigned to the emergency information data record at a later point in time in order to form a training data record for training the trained prediction function. Method according to one of the preceding claims, characterized in that the retrieved emergency information data set, in particular the occupant impact sensor data, are evaluated on a display device of a computing device (28) of the rescue service for an optical visualization of the impact on the at least one occupant, in particular for a representation of the processes in the interior (2) relating to the at least one occupant. Motor vehicle (1), comprising a device for carrying out a method according to one of the Claims 1 until 11 control device (19) and the storage means (20). Emergency information system (23), comprising a motor vehicle (1) according to Claim 14 and an external server (25).

Images