Methods for preconditioning protective measures in a semi-autonomous motor vehicle and a simulation device

Abstract

Method for preconditioning protective measures for at least one occupant (10) in an interior (12) of at least a semi-autonomous motor vehicle (14), in which at least one simulation of a potential accident situation is simulated by means of a simulation device (16) and, depending on the simulation, at least one signal for controlling a control device (18) of the motor vehicle (14) is generated and transmitted to the control device (18),wherein, by means of a detection device (20), predefined parameters of protection-relevant control devices (18) within the interior (12) are first detected and transmitted to the simulation device (16), and depending on the parameters, a simulation is carried out to calculate specific protective measures for the protection of the at least one occupant (10) in the event of an actual accident situation, and wherein, in the actual event of an accident situation, one of the protective measures is loaded from the database (22), and depending on the loaded protective measures, the at least one signal for controlling the at least one control device (18) is generated and transmitted to the control device (18), characterized in that, by means of an artificial intelligence arranged in the simulation device (16), a clustering of detected data from the simulations is carried out.in which, based on previously known parameters and data from the database (22), it is calculated which protective measures and output signals are suitable for potential accident situations, whereby parameter values ​​of the parameters are also taken into account by means of the simulation device (16) to determine efficient specific protective measures.

Description

[0001] The invention relates to a method for preconditioning protective measures for at least one occupant in an interior of an at least partially autonomous motor vehicle according to the preamble of patent claim 1. Furthermore, the invention relates to a simulation device.

[0002] In a use case of customer fleet data to improve passenger safety, preconditioning measures are already being implemented before an accident or crash. For example, windows are automatically closed, seatbelts are tightened, etc. However, these measures are relatively standardized and not adapted to specific situations.

[0003] DE 10 2019 205 006 A1 discloses a computer-implemented method for operating safety systems of a vehicle. This method comprises selecting an accident scenario for the vehicle, analyzing the effects of this accident scenario on the vehicle's occupants, simulating at least the positions and postures of the occupants at the time of the accident, depending on the determined effects, determining the safety systems to be activated in the accident scenario to protect the occupants, and storing the accident scenario and the selected safety systems in the vehicle's memory.

[0004] EP 1 592 584 B1 already discloses a method for presetting a vehicle occupant protection system, which comprises a plurality of adaptive protection system components, each assigned to a seat in the vehicle, and is preset using occupant-related data. It is provided that, for at least one seat, the identity of the occupant occupying the seat is determined, a safety profile associated with the identified person and stored in a vehicle-mounted memory is retrieved, and the protection system components assigned to the seat are set based on the safety profile.

[0005] DE 10 2009 012 407 B3, DE 103 17 211 A1 and WO 2017 / 102740 A1 are respective inventions of the prior art relating to the method according to the invention for preconditioning protective measures in an at least partially autonomous motor vehicle.

[0006] The object of the invention is to provide a method for preconditioning protective measures and a simulation device by means of which at least one situational protective measure that can be adapted to an occupant of a motor vehicle is carried out.

[0007] This object is achieved by means of a method having the features of patent claim 1 and by means of a simulation device according to the invention. Advantageous refinements of the method according to the invention are to be regarded as advantageous embodiments of the simulation device according to the invention, wherein the simulation device in particular comprises means for carrying out the method steps. Advantageous refinements with favorable developments of the invention are also the subject of the dependent claims.

[0008] A first aspect of the invention relates to a method for preconditioning protective measures for at least one occupant in the interior of an at least partially autonomous motor vehicle, in particular a passenger car, in which at least one simulation of a potential accident situation is simulated by means of a simulation device. The simulation device is in this case, for example, an electronic computing device with all the necessary components for carrying out at least one simulation, wherein software, algorithms, or artificial intelligence can be used for this purpose. All input signals and parameters are accordingly detected by the sensor device and transmitted to the electronic computing device. Subsequently, output signals are generated by means of a software system assigned to the electronic computing device, which are intended to trigger at least one protective measure for the occupant in the accident situation.The system envisages that, depending on the simulation, at least one signal, namely an output signal, is generated to control a control device of the motor vehicle and transmitted to the corresponding control device. Preconditioning protective measures using the simulation device can help minimize the risk to people and property and improve the effectiveness of protective measures by identifying and correcting potential weaknesses and problems as quickly as possible before they occur during an accident.For example, an accident situation can be detected before it occurs using at least one sensor device, and at least one input signal can be transmitted to the simulation device, whereby a simulation of the accident situation is carried out within a predetermined time window and, for example, tightening a belt and changing the occupant's seating position are selected as possible protective measures. Both the belt and the seating position can be at least partially automatically controlled or adjusted by means of a respective control device that receives the control signal, whereby at least one protective measure for the occupant is implemented before the accident situation occurs.

[0009] In order to achieve the object of the invention and thus implement at least one situational protective measure adapted to an occupant of a motor vehicle by means of the method, the invention provides that predetermined parameters of protection-relevant control devices within the interior are first detected by means of a detection device. The detection device is used to detect data and information from the protection-relevant control devices for controlling, for example, a seat system and / or a steering wheel position and / or a seat belt tensioning and / or for opening and closing windows, etc. Thus, the parameters of the respective control devices are detected and transmitted to the simulation device as input signals.Depending on the parameters, a simulation is carried out to calculate specific protective measures to protect at least one occupant in the event of an actual accident situation. The preconditioning of protective measures using the simulation device refers to the process in which protective measures are tested and optimized using computer models and simulations before an actual event occurs. The simulation device can be used to test scenarios and evaluate the effectiveness of protective measures before the accident situation occurs. The preconditioning of protective measures using the simulation device using the specified parameters as input signals thus refers to a process in which the protective measures are developed and optimized using computer models and simulations of the electronic computing device, using the various specified parameters as input signals.The protective measures implemented using the simulation device with the parameters as input signals can help minimize the risk to people and vehicles and improve the effectiveness of protective measures by generating realistic scenarios and evaluating reactions to various accident situations and other possible events. The output signals of the simulation device are used to control the control devices or other devices. The control devices receive the electrical or electronic signals and, depending on the signals, perform at least one corresponding adjustment to protect the occupant.

[0010] The simulation facility takes into account the parameter values ​​to determine more effective, specific protective measures. If the parameter values ​​of individual occupants vary significantly, additional measures specifically necessary for the occupant's protection are calculated based on the parameter values. This allows for a better response to the occupant's specific physical characteristics.

[0011] Finally, it is planned that artificial intelligence (AI) installed in the simulation facility will be used to cluster the data collected during the simulations. The AI-based learning process will use the already known parameters and data from the database to calculate which input signals correspond to which protective measures, and which protective measures and output signals correspond to potential accident situations. By applying AI, the required protective measures, particularly those adapted to the situation, can be applied more quickly and efficiently. Clustering will utilize a machine learning method to organize data points of the respective vehicle into groups. Similarities in the data will be used to identify groups that are as homogeneous as possible, particularly vehicle groups of the same model.Likewise, potential accident situations should be grouped into groups or clusters. Using clustering or a clustering procedure, the already known situations are classified into clusters, and the cluster corresponding to the current vehicle or situation is applied. Examples of measures when selecting a cluster include, depending on the vehicle model, setting partial airbag deployment, placing seats upright, folding up tables, sucking in objects and / or collecting them with a super magnet and / or a gripper arm, injecting super foam, restraining hands and / or limbs, tightening a seatbelt, illuminating the interior with a blue light, rolling up windows, etc.

[0012] In other words, the parameters are recorded and, using simulation, parameters are calculated which represent a more advantageous position of the respective protection-relevant control devices for the occupant in the event of an accident. This means that parameters are calculated which are then transmitted to the respective control devices in order to better protect the occupant in the event of an accident. In numerical mathematics, preconditioning refers to the technique by which a problem, namely the possible accident situation, is transformed in such a way that the solution, namely the protection of the occupant by adjusting the respective control devices by calculated parameters, is retained. Accordingly, the protective measure is seen as the solution, with the recording device recording the parameters required for the calculation. Since all occupants are different occupants with different body sizes, statures, etc.It is also intended that the occupant's measurements are recorded by the recording device and taken into account in the calculation of protective measures.

[0013] Above all, in more highly automated driving states (Levels 3-5), the occupant should have the opportunity to engage in non-driving activities during a journey. These activities could include reading a book, using a notebook, knitting, etc. In such a case, a driver as the occupant and a front passenger seat may also face a rear seat of the vehicle, similar to a four-seater seat in an ICE train. Thus, the basic assumption for accidents in which the driver has both hands on the steering wheel and is seated facing the direction of travel no longer applies. Therefore, protective measures must be taken to protect the driver and the other occupants, even if they are using potentially dangerous objects such as a notebook while driving or even sleeping in a comfortable, non-upright position.In addition to the condition of the vehicle interior, the type of anticipated accident or collision also significantly influences how the at least partially autonomous vehicle should be optimally preconditioned for the potential accident situation, such as a crash, in order to mitigate the consequences of the actual accident. For example, in a head-on collision, different forces act on the vehicle occupants and objects than in a side collision. Due to the large number of parameters, such as the number of passengers, seating positions, unsecured objects in the interior, the vehicle body, the load, etc., it is not possible, for cost and time reasons, to provide measures that optimally protect the occupants and others for all scenarios.

[0014] In an advantageous embodiment of the invention, it is provided that, depending on the parameters, a plurality of specific protective measures for protecting the occupant are simulated and stored in a database arranged in the simulation device. The protective measures already calculated and, if applicable, implemented are saved in the database for later use, whereby the protective measures can be implemented more quickly without having to run a simulation again. Occupant-specific protective measures can also be saved, which are applied depending on the occupant sitting in the vehicle. For this purpose, the occupant is recorded, and data from the database relating to the occupant is applied to the simulations. The calculated protective measures are then saved with a link to the occupant and reapplied for future potential accident situations.

[0015] In another advantageous embodiment of the invention, in the actual event of an accident situation, one of the protective measures is loaded from the database and, depending on the loaded protective measures, at least one signal for controlling the at least one control device is generated and transmitted to the control device. The adjustments in the interior required by the protective measures are thus carried out by the control devices, which are adjusted depending on the protective measure in such a way that the occupant is helped as efficiently as possible in the actual accident situation. The aim is thus to control a protection system in the interior by means of which several assigned adaptive components are adjusted in relation to the occupant in such a way that the occupant's safety is increased.

[0016] In yet another advantageous embodiment of the invention, it is provided that a trigger signal for starting the simulation is transmitted to the simulation device by means of a sensor device of the at least partially autonomous motor vehicle. In this case, it is provided that the sensor device detects, for example, parameters relating to weather conditions in the vicinity of the motor vehicle and a distance to the vehicle in front, whereby exceeding a critical value of the detected parameters triggers the trigger signal. This signal is thus transmitted to the simulation device, and the parameters are used as input signals for the simulation. Potential accident situations detected by the sensor device include, for example, large expected accelerations of passengers, short distances to surrounding vehicles and objects, as well as data from ESP / ABS.In addition, all sensors and components arranged in the sensor device or coupled to the sensor device can be used, for example interior cameras, microphones, temperature sensors, seat positions, weight sensors, etc.

[0017] In yet another advantageous embodiment of the invention, the simulation device is designed as an electronic computing device external to the vehicle. In particular, this is intended to enable the coupling of a large number of motor vehicles with the electronic computing device external to the vehicle, provided, for example, as a server with a cloud, whereby data from a fleet can be recorded and stored. This allows already known protective measures and simulations to be provided for all vehicles in the fleet.Example data that can be sent from the motor vehicle to the vehicle-external electronic computing device or to a backend of the server to describe the situation are, for example, point clouds or objects already detected in a room, which can be compared (FEM), classified, detected objects from image data, body postures of the occupants from image data and pressure sensors in the seat, seat settings, belt tension, seat occupancy, detection of the situation from acoustic signals, crash predictions for a frontal crash, side crash, etc.

[0018] An embodiment of the invention has also proven advantageous in which the simulation device retrieves data from previously recorded accident situations from a database and feeds this data into at least one simulation to calculate specific protective measures. Similarity metrics, such as weighted correspondences of the number of typed objects, are to be applied. In particular, the fleet data is to be applied to all vehicles in the fleet.

[0019] In addition, it has proven advantageous to search the database for known specific protective measures based on the recorded parameters before a new simulation. The previously calculated and, if applicable, implemented protective measures from the entire fleet are saved in the database for later use, allowing the protective measures to be implemented more quickly without having to repeat a simulation. In particular, this also allows for an evaluation of the protective measures, which also results in particularly efficient protective measures receiving a higher rating. Depending on the evaluation, these can then be applied to the potential accident situation.

[0020] A second aspect of the invention relates to a simulation device with all electronic components, modules, circuits and links for carrying out a method for preconditioning protective measures for at least one occupant in an interior of an at least partially autonomous motor vehicle according to the first aspect.

[0021] In other words, the invention provides for the use of data-supported simulations. In the event of an anticipated or suspected collision, or even in the event of very sudden braking maneuvers involving large forces, an individual simulation of the event is to be carried out and the optimal protective measures to mitigate the consequences of the accident are to be calculated. The simulation is carried out shortly before the collision, for example, on an electronic computing device external to the vehicle. In order to limit the solution space for the simulation and to solve the problem as effectively as possible in a limited time, a database of previously occurred accident situations or crashes is used. Similar situations are sought from this database. The measures that were particularly effective in these situations are used to simulate the current accident situation.Various procedural steps can be used here:. - Collecting data for relevant scenarios via crash tests and real crashes from a customer fleet. - Simulating scenarios with different parameter values ​​to determine the optimal measures to mitigate the consequences of the accident for the occupants. - In real-world operation, a vehicle in a critical situation queries the backend with its setup, for example, number of passengers, seating positions, activities, etc., and requests an optimal set of protective measures for the situation. The backend searches for known situations and their associated measures, for example, "head-on collision with three occupants, one of whom is using a laptop." N similar situations are found, and the M measures that were particularly helpful in the situation are applied in the next process step. - A simulation of all M measures is performed using the request from the current vehicle. The result is a set of N presumably particularly effective protective measures, for example, adjusting the seat position to the specific situation of the vehicle. - Returning a suitable set of protective measures to the vehicle in which they will be implemented to successfully avert the critical situation.

[0022] Further features of the invention emerge from the claims, the figures, and the description of the figures. The features and combinations of features mentioned above in the description, as well as the features and combinations of features mentioned below in the description of the figures and / or shown alone in the figures, can be used not only in the respective specified combination, but also in other combinations or on their own.

[0023] The invention will now be explained in more detail using a preferred embodiment and with reference to the drawing. It shows: Fig. 1 A picture diagram illustrating a method according to the invention for preconditioning protective measures for at least one occupant in an interior of an at least partially autonomous motor vehicle.

[0024] The only figure Fig. 1 shows an image diagram to illustrate a method according to the invention for preconditioning protective measures for at least one occupant 10 in an interior 12, in particular on a vehicle seat 13 of an at least partially autonomous motor vehicle 14. A simulation device 16 with all electronic components, modules, circuits and links for carrying out the method for preconditioning protective measures is to be used.

[0025] In the method for preconditioning protective measures for the at least one occupant, at least one simulation of a potential accident situation is to be simulated by means of the simulation device 16 and, depending on the simulation, at least one signal for controlling a control device 18 of the motor vehicle 14 is to be generated and transmitted to the control device 18. However, by means of a detection device 20, predetermined parameters P of protection-relevant control devices 18 within the interior 12 are first detected and transmitted to the simulation device 16 and, depending on the parameters P, a simulation is carried out to calculate specific protective measures for protecting the at least one occupant 10 in the event of an actual accident situation.Protection-relevant control devices 18 include, for example, the control devices 18 of seat systems or belts, which can be controlled in such a way that the occupant 10 is at least partially protected in the event of an actual accident situation, for example, by sitting down and tightening the belt. Sitting down and tightening the belt are therefore considered protective measures. Fig. 1, such a first protective measure S1 is the turning or adjustment of the vehicle seat 13 facing rearward in the vehicle longitudinal direction x forward in the vehicle longitudinal direction x in order to protect the occupant 10 sitting thereon in the actual accident situation.

[0026] Depending on the parameters P, a plurality of specific protective measures for protecting the occupant 10 are to be simulated and stored in a database 22 arranged in the simulation device 16 in order to be able to access the protective measures more quickly later without having to repeat a simulation. Thus, in the actual event of an accident situation, one of the protective measures can be loaded from the database 22, and depending on the loaded protective measures, the at least one signal for controlling the at least one control device 18 can be generated and transmitted to the control device 18, thereby providing a faster protective measure.

[0027] Likewise, a trigger signal can be transmitted to the simulation device 16 as an input signal for starting the simulation by means of a sensor device 24 of the at least partially autonomous motor vehicle 14. The sensor device detects, for example, parameters P relating to weather conditions in the surroundings of the motor vehicle and a distance to the preceding vehicle, whereby exceeding a critical value of the detected parameters P triggers the trigger signal. This signal is thus transmitted to the simulation device, and the parameters P are used as input signals for the simulation.

[0028] The simulation device 16 can also be designed as an electronic computing device 26 external to the vehicle, wherein both the electronic computing device 26 external to the vehicle and the sensor device 24 or the motor vehicle 14 have data modules for transmitting and capturing data. In particular, the electronic computing device 26 external to the vehicle can be provided as a cloud server. It is provided that a plurality of motor vehicles 14 are coupled to the electronic computing device 26 external to the vehicle in order to collect and store as much data as possible. By means of the simulation device 16, in particular by means of the electronic computing device 26 external to the vehicle, data from previously recorded accident situations can be obtained from a provided database 28 and fed into the at least one simulation for calculating specific protective measures.

[0029] In particular, the simulation device 16 is intended to consider parameter values ​​of the parameter P to determine more efficient specific protective measures. Before a new simulation and depending on the recorded parameters P, already known specific protective measures are also to be searched in the database 28. Finally, a clustering of recorded data from the simulations is to be performed using an artificial intelligence system installed in the simulation device 16. List of reference symbols 10 inmates 12 Interior 13 Vehicle seat 14 motor vehicles 16 Simulation facility 18 Control device 20 Recording device 22 Database 24 Sensor device 26 vehicle-external electronic computing device P parameters S1 first protective measure X Vehicle longitudinal direction

Claims

[1] Method for preconditioning protective measures for at least one occupant (10) in an interior (12) of an at least partially autonomous motor vehicle (14), in which at least one simulation of a potential accident situation is simulated by means of a simulation device (16) and, depending on the simulation, at least one signal for controlling a control device (18) of the motor vehicle (14) is generated and transmitted to the control device (18),wherein, by means of a detection device (20), initially predetermined parameters of protection-relevant control devices (18) within the interior (12) are detected and transmitted to the simulation device (16), and depending on the parameters, a simulation is carried out to calculate specific protective measures for protecting the at least one occupant (10) in the event of an actual accident situation, and wherein, in the actual case of an accident situation, one of the protective measures is loaded from the database (22), and depending on the loaded protective measures, the at least one signal for controlling the at least one control device (18) is generated and transmitted to the control device (18), characterized bythat a clustering of recorded data from the simulations is carried out by means of an artificial intelligence arranged in the simulation device (16), in which clustering is carried out from already known parameters and data from the database (22) which protective measures and output signals are suitable for potential accident situations, wherein by means of the simulation device (16) parameter characteristics of the parameters are also taken into account for determining efficient specific protective measures. [2] Method according to claim 1, characterized by that, depending on the parameters, a plurality of specific protective measures for protecting the occupant (10) are simulated and stored in a database (22) arranged in the simulation device (16). [3] Method according to one of the preceding claims 1 or 2, characterized bythat a trigger signal for starting the simulation is transmitted to the simulation device (16) by means of a sensor device (24) of the at least partially autonomous motor vehicle (14). [4] Method according to one of the preceding claims, characterized by that the simulation device (16) is designed as an electronic computing device (26) external to the vehicle. [5] Method according to one of the preceding claims, characterized by that by means of the simulation device (16) data of already recorded accident situations are obtained from a database (28) and fed into the at least one simulation for calculating specific protective measures. [6] Method according to one of the preceding claims 2 to 5, characterized by that before a new simulation and depending on the parameters recorded, already known specific protective measures are searched in the database (28). [7] Simulation device (16) with all electronic components, modules, circuits and links for carrying out a method for preconditioning protective measures for at least one occupant (10) in an interior (12) of an at least partially autonomous motor vehicle (14) according to one of the preceding claims 1 to 6.

Images