System for controlling the braking behavior of a vehicle

Abstract

System for controlling the braking behavior of a vehicle, comprising a central or decentralized control unit (13, 33) which is connected to a plurality of vehicle sensors (5, 7, 9, 11), one of which sensor (5) is coupled to a brake pedal (3) for determining its position or movement, characterized in that the control unit (13, 33) is equipped to detect an actual blockage of the brake pedal (3) or to predict an impending blockage by means of: - a force sensor arranged on a brake pedal plate for checking the proportionality of the actuation force to the brake pedal travel, - an accelerator pedal travel sensor for detecting unintentional actuation, - an ultrasonic sensor arranged in the footwell of the vehicle (1) for detecting foreign objects behind or next to the brake pedal (3) via distance detection,- a camera arranged in the footwell of the vehicle (1) for detecting foreign objects behind or next to the brake pedal (3) by object identification in an image of the footwell, - a force sensor arranged on the floor of the footwell of the vehicle (1) behind the brake pedal (3) for detecting foreign objects behind the brake pedal (3) by means of the brake pedal force transmitted from it to the floor of the footwell, - an interior microphone arranged near the brake pedal (3) for recording noises produced by foreign objects coming into contact with the brake pedal (3), - a seat mat or a force sensor in a seat cushion for determining a force transfer from the seat cushion to the backrest of the driver's vehicle seat when the brake pedal (3) is pressed, - another interior camera for detecting a blocked brake pedal (3) based on the driver's facial expression, - another interior microphone for detecting the blocked brake pedal (3) by means of exclamations from the driver,- a force sensor on the steering wheel for detecting the blocked brake pedal (3) by the driver gripping the steering wheel more tightly; - a pulse sensor for detecting the blocked brake pedal (3) by an increase in the driver's pulse; - at least one environment sensing sensor (11) for detecting the blocked brake pedal (3) by a reduction in the distance of the vehicle (1) to other vehicles or obstacles, wherein the control unit (13, 33) has an artificial intelligence (15) which is trained once in a test vehicle using the sensor signals obtained during a test drive, whereby a predetermined number of tuples (21) are generated, each comprising an explanatory feature (21a) formed by measurement data from the sensors (5, 7, 9, 11) and an explained feature (21b) in the form of the blocking state of the brake pedal (3) prevailing at the time of measurement; wherein all generated tuples (21) are divided into a training set (25) and a test set (27)whose sets do not overlap, are divided, with which a mapping function for the prediction of the blocking states of the brake pedal (3) is generated using one or more supervised learning methods, wherein the prediction method to be applied of the artificial intelligence model (15) is determined on the basis of a degree of deviation of a predicted unknown tuple (29) from the test set (27), which is implemented in the control unit (13, 33) of a production vehicle (1), in which the sensor data transmitted by the sensors (5, 7, 9, 11) to the control unit (13, 33) are processed in such a way that they are used by the implemented artificial intelligence model (15) for predicting the brake pedal state in order to adapt the brake pedal tension in order to detect the driver's braking request even in the case of partial blocking of the brake pedal travel.

Description

The invention relates to a system for controlling a braking behavior of a vehicle, comprising a central control unit which is connected to a multiplicity of vehicle sensors, of which a sensor for determining a position or movement of a brake pedal is coupled to the latter.DE 10 2008 029 311 A1 discloses a brake system for a motor vehicle and a method for controlling a vehicle brake, wherein the brake pedal unit comprises a sensor which has its own intelligence in the form of an electronic arithmetic unit which calculates actuating signals for brake actuators directly. Furthermore, the brake system comprises a central electronic control unit which, on the basis of the signals of the sensor in the brake pedal unit or of a further sensor, generates actuating signals for actuators. The actuators, which also have their own intelligence in the form of an electronic arithmetic unit, check the validity of the signals output by the central control unit. If the control signal output by the central control unit is valid, it is executed by the actuators. If the actuating signal of the central control unit is not valid, the actuators are actuated on the basis of the actuating signal calculated by the sensor of the brake pedal unit.DE 10 2021 117 917 A1 discloses a method for assisting a driver of a motor vehicle, which has an interior with a driver's seat, a camera and a data processing device. The camera is used to generate image data which depict a part of the interior in the region of the driver's seat in which the position of at least a part of an upper half of the body of the driver is to be expected. This image data is evaluated by the data processing device to the effect that a physical reaction of the driver is detected, which is assigned to a braking request.DE 10 2021 126 461 A1 discloses a method for detecting a resistance at a brake pedal of a brake system, which brake pedal is pulled by an actuator, in particular a brake booster, in which the freeness of the brake pedal is determined during operation of the actuator. The actuator for actuating the brake pedal is controlled as a function of the resistance determined.DE 10 2019 130 309 A1 discloses a pedal system for a foot-executable command input for a motor vehicle, comprising a control unit which generates a control signal for actuating an actuator as a function of a travel sensor output signal determined by a travel sensor.It is an object of the invention to specify a system for controlling a braking behavior of a vehicle which, despite the detected or predictable blocking of the brake pedal, allows the driver to continue to influence the vehicle deceleration.The invention results from the features of the independent claims. Advantageous refinements and refinements are the subject matter of the dependent claims. Further features, possible applications and advantages of the invention are evident from the following description and from the explanation of exemplary embodiments of the invention which are illustrated in the figures.The object is achieved by the subject matter of claim 1.In the system explained at the beginning for controlling a braking behavior of a series vehicle, comprising a central or decentralised control device which is connected to a plurality of vehicle sensors, of which a sensor for determining a position or movement of a brake pedal is coupled to the latter, the control device of a series vehicle is used for detecting a blockage of the brake pedal which is currently occurring or for predicting an imminent blockagea force sensor arranged on a brake pedal plate for checking the proportionality of the actuating force to the brake pedal travel,an accelerator pedal travel sensor for detecting an unintentional actuation,an ultrasonic sensor arranged in the foot well of the vehicle for detecting foreign objects behind or next to the brake pedal via a distance detection,a camera arranged in the foot well of the vehicle for detecting foreign objects behind or next to the brake pedal by object identification on an image of the foot well,a force sensor arranged on the floor of the foot of the vehicle behind the brake pedal for detecting foreign objects behind the brake pedal by the brake pedal force transmitted from the latter to the floor of the foot,an interior microphone arranged near the brake pedal for receiving noises which generate foreign bodies which come into contact with the brake pedal,a seat mat or a force sensor in a seat cushion for determining a force displacement from the seat cushion to the backrest of the vehicle seat of the driver when the brake pedal is actuated,a further interior camera for detecting a blocked brake pedal on the facial expression of the driver,a further interior microphone for detecting the blocked brake pedal by calling out the driver,a force sensor on the steering wheel for detecting the blocked brake pedal by the driver clamping the steering wheel more strongly around,a pulse sensor for detecting the blocked brake pedal by a pulse increase of the driver,at least one environment detection sensor for detecting the blocked brake pedal by reducing the distance of the vehicle from other vehicles or obstacleswherein the control device has an artificial intelligence which is trained once in a test vehicle on the basis of the sensor signals determined during a test run, wherein a predefined number of tuple are generated which each comprise an explanatory feature which is formed by measurement data of the sensors and an explained feature in the form of the blocking state of the brake pedal which prevails at the time of the measurement, wherein all generated tuple are divided into a training set and a test set whose sets have no overlaps, with which a mapping function for predicting the blocking states of the brake pedal is generated by means of one or more methods of the monitored learning, wherein the prediction method of the model of the artificial intelligence to be applied is determined on the basis of a degree of the deviation of a predicted unknown tuple from the test set, wherein the prediction method of the model of the artificial intelligence is determined on the basis of a degree of the deviation of a predicted unknown tuple from the test set, which is implemented in the control device of the series vehicle in which the sensor data transmitted from the sensors to the control device are processed in such a way that they are used by the implemented model of the artificial intelligence for predicting the brake pedal state for adapting the brake pedal tension for detecting the braking request of the driver even in the case of partial blocking of the brake pedal travel.A model of artificial intelligence is to be understood below as a software function which can make classifications and value predictions on the basis of input information. This mapping function is trained using methods of machine learning.A brake pedal application is to be understood as an adjustment of the brake deceleration power, which is meterable for the driver, over the usable brake pedal travel.A brake pedal blockage describes a movement restriction of the brake pedal triggered by a foreign body or mechanical component defects.The evaluation of the plurality of sensor signals in a model of artificial intelligence permits reliable detection of an existing blockage of the brake pedal and also the prediction of an imminent blockage. Not only signals from a brake pedal sensor are used here, but additionally signals from sensors are also included, which detect the brake pedal environment, the driver and the vehicle environment. This increases the accuracy of the detection. In addition to the information about the blocking of the brake pedal, the recognition function of the model of the artificial intelligence also provides information about from which position the brake pedal is blocked.Due to the high sensitivity of the detection of the blockage of the brake pedal, objects of different size and material quality can be detected.The method for detecting brake pedal blockages is divided into the individual steps; generation of training data, training of the model of the artificial intelligence and application of the model in the vehicle.The information about the blocking position is used to dynamically adapt the brake pedal application in such a way that the driver can request the full deceleration performance of the brake despite the restricted brake pedal travel. If there is a blockage of the zero position of the brake pedal, then the brake pedal application is adapted in such a way that the vehicle can no longer be braked unintentionally. The adaptation of the brake pedal application makes it possible for the driver to continue to meter the deceleration power within the remaining travel of the brake pedal.In one embodiment, training data is generated with a simulation method for training the model of the artificial intelligence, which simulation method summarizes the physical relationships between the behavior of the driver, the behavior of the brake pedal and the properties of the foreign body into a software function, wherein a real driving maneuver of the vehicle is simulated with the simulation and data of virtual sensors are generated. As a result, more data can be provided for training the model of the artificial intelligence in the shortest time and with less resource outlay.In a further embodiment, a method of linear and polynomial regression and / or the nearest neighbor classification and / or neural networks can be used for training the model of the artificial intelligence. All of these methods make use of a mapping function for predicting the blocking states of the brake pedal underCreated with the aid of the training data and mathematical optimization methods. The method, which calculates the best possible approximation to an ideal mapping function, is determined with metrics that calculate the degree of deviation of the respective model of the artificial intelligence when predicting unknown tuple from the test block of the sensor data.In a further refinement, the control unit is designed to execute a calculation of the brake pedal state cyclically at a high frequency. As a result, a change in the brake pedal state can be detected very quickly and a corresponding substitute reaction can be triggered.In a further embodiment, the control device classifies the brake pedal and distinguishes between "brake pedal blocked" and "brake pedal unblocked" in the classification, wherein, if the brake pedal has been classified as blocked, it is again differentiated whether the blockage is present at the end stop or at the zero position of the brake pedal. An end stop is detected, for example, when an object is behind the brake pedal. A zero position is concluded when the object is on the brake pedal. Regardless of the type of blockage, the clamping of the deceleration power is adapted to the remaining free-wheeling residual travel of the brake pedal.In a further embodiment, a transition between different brake pedal voltages occurs abruptly when the brake pedal is released or in a time ramp when the brake pedal is permanently actuated. As a result, transitions to a changed brake pedal application remain controllable for the driver.In a further embodiment, the control device is connected to a warning device for outputting a warning when a critical state of the brake pedal is determined. The driver of the vehicle is thus warned in an predictive manner even before a dangerous driving situation occurs. He can place the vehicle in a safe location in good time in order to remove any object in the vicinity of the pedal.Further advantages, features and details are evident from the following description, in which at least one exemplary embodiment is described in detail. Features described can form the subject matter of the invention alone or in any meaningful combination, optionally also independently of the claims, and can in particular additionally also be the subject matter of one or more separate application / s.The following are shown: FIG. 1 shows a schematic illustration of the system (sensor+actuator) according to the prior art, FIG. 2 shows an exemplary embodiment for generating training data for the model of the artificial intelligence, FIG. 3 shows an exemplary embodiment for training the model of the artificial intelligence, FIG. 4 shows an exemplary embodiment for applying the trained model of the artificial intelligence in the vehicle, FIG. 5 shows exemplary embodiments for adapting the brake pedal application.FIG. 1 shows a schematic illustration of the system according to the invention, which is arranged in a vehicle. The vehicle 1 comprises a plurality of sensors, of which only some are shown by way of example in the drawing. A brake pedal sensor 5 is arranged on a brake pedal 3 and a foot space sensor 7 is arranged in the vicinity of the brake pedal 3. Table 1 contains a detailed summary of sensors suitable for detecting brake pedal blockage. These sensors 5, 7, like a driver monitoring sensor 9 and a vehicle environment sensor 11, are connected to a central control unit 13 of the vehicle 1, which comprises a model of artificial intelligence 15 which evaluates the sensor signals with regard to a blockage of the brake pedal 3. Depending on the result provided by the model of the artificial intelligence 15, the central control unit 13 controls a braking device 17 of the vehicle 1. Advantageously, the sensors 5, 7, 9, 11 and the brake device 17 communicate with one another via a data bus 19 of the vehicle 1. Table 1 Table 1Brake pedal Brake PedalPedal travel sensorBremspedalBrake Pedal TravelAbsolute brake pedal travel remains continuously under the control of driving cycles or over the occurrence of characteristic brake travel gradients, occurrence of characteristic actuation frequencies (panic repeated depression of the brake pedalForce Sensor OnForce acting on the load.Brake pedal force notBrake Pedal PlateThe brake pedal plate acts as a brake pedal platemore proportional to the travel distancePedal travel sensorFahrpedalAccelerator Pedal Travel PathIf the accelerator pedal is unintentionally co-actuated with the brake pedal being actuated, accelerator pedal travel >0% and characteristic acceleration gradients occurFoot wellUltrasonic sensorDistance to Floor Space SupportUltrasonic sensor detects ultrasonic force behind or beside the brake pedalCameraObject identification on a video image of the foot wellIn the image, brake signals are detected behind and beside the brake pedalForce Sensor on Floor Rear of the Floor of the VehicleForce acting on the floor of the foot wellIf a vehicle body is located behind the vehicle body, then the vehicle body transmits the force of gravity to the floor of the footInterior microphoneNoise in Foot SpaceUpon contact of the substrate with the substrate, a characteristic noise is producedDriver DriverSeat mat / force sensor in seat cushionForce acting on the load cellWhen the driver steps on the brake pedal sharply, the backrest is pressed and the seat cushion is relieved of loadInterior CameraRecognition of the Mimic of the Driver on Video ImageThe mimic of the driver can be used to identify when the driver is going to a stop in the brake pedal or in the panic deviceInterior microphoneInterior Noise, Utterance of DriverCharacteristic sounds / excretions of the callee, such as "Ah!","Hergotts AGG", which are associated with technical defect instructionsForce Sensor on the Load Bearing DeviceForce by which the planet carrier is surroundedWhen the driver is engaged by a locked brake pedal, he encloses the steering wheel with a higher forcePulse Sensor / Fitness BraceletPulse of the driverWhen the driver is engaged by a brake pedal blocked inIf the panic device is going to be, then its pulse will rise abruptlyVehicle environmentEnvironment Detection with Light Beam, Camera, LIDAR sensor, Ultrasonic sensorDistance to other vehicles and vehicles, for example as output signal of a passive distance control temperatureWhen the brake pedal is blocked, the travel distance may be smaller than is expected in the respective vehicle range in order not to fall below the distance between other vehicles and vehicle rangesTo identify the blockage, the model of the artificial intelligence 15 must be trained using various machine learning methods before use in the vehicle 1. The training is divided into the steps "generation of training data", "training of the model" and "application of the model".For training the model of the artificial intelligence 15, the method of supervised learning (supervised learning) is used. For this purpose, a set of tuple 21 is first generated, as is shown in FIG. 2. Each tuple 21 consists of an explanatory feature 21 a(feature) and an explanatory feature 21 b(label). The features 21 aare the measurement data of the sensors 5, 7, 9, 11 for which the blocking state of the brake pedal 3 prevailing at the time of the measurement is stored as label 21 bin a tuple 21. Vehicle tests are carried out to generate representative tuple 21. In this case, typical blocking states of the brake pedal 3 are specifically brought about by different objects being positioned one after the other in the foot space in such a way that a blocking of the brake pedal 3 occurs. At the same time, typical driving maneuvers are carried out with the prepared vehicle on a shut-off test site and measurement data are recorded. The identical driving maneuvers are also carried out and measured using a free-wheeling brake pedal 3. In order to best image the situations typically occurring in reality, both the driving maneuvers and the properties of the foreign bodies are uniformly distributed within defined limits. Table 2 shows possible foreign bodies blocking the brake pedal 3. Table 2 Table 2Small flat shapeFoot matTissue PackPortemonnaie, mobile telephoneLarge, thick shapeWarning vest,Flip-flop shoes.ice scratches, ice scratches, and ice scratches,InserterEmpty BottleHandbeeen, full bottle of tobacco, umbrellaTo increase the efficiency in data generation, simulation methods are also used in which the relevant physical relationships between the driver 23, the behavior of the brake pedal 3 and the properties of the foreign body are modeled in a software function. Real driving maneuvers in the test vehicle are then simulated and data of virtual sensors are generated.According to FIG. 3 a, the set of all tuple 21 is divided into a training set 25 and a test set 27. The two sets must be disjunctive so that an objective assessment of the performance of the model of the artificial intelligence 15 can be carried out on the basis of the test set 27.Should training data be present in the test set 27 by mistake, it may happen that the performance of the model is overestimated. The model of the artificial intelligence 15 could not have mapped the general relationship between the sensor data and the brake pedal blockage, for example, but could only correctly predict the learned training data and thus only known states.The real performance of the model of the artificial intelligence 15 could be poorer than expected in the case of unknown sensor data, for example in the case of deviating driving maneuvers or in the case of new geometries or material properties of the foreign bodies. This behavior of the model of the artificial intelligence 15 is referred to as an "over-fitting" and can be prevented by the correct division into training and test sets 25, 27.FIG. 3 shows an exemplary embodiment for training the model of the artificial intelligence 15. To generate a suitable mapping function for predicting the blocking states of the brake pedal 3, a plurality of methods of supervised learning are available. These include, for example, linear and polynomial regression, "nearest neighbor classification" or artificial neural networks (ANN). These methods have in common that they calculate an optimum approximation of the ideal mapping function by means of the training data of the training set 25 and mathematical optimization methods (FIG. 3 b).In order to determine the most powerful of the selected methods for the present detection of the blocking state of the brake pedal 3 and the sensors 5, 7, 9, 11 to be selected for this purpose, all methods to be checked with respectively different configurations (hyperparameters of the method) are examined and validated. Metrics which calculate the degree of deviation of the model of the artificial intelligence 15 when predicting unknown tuple 29 from the test set 27 serve as the objective evaluation criterion (FIG. 3 c ).The previously one-time and offline trained model of the artificial intelligence 15 is implemented as a software function on the central control device 13 of the vehicle 1 after successful validation in FIG. 3 c. Alternatively, this software function can also be implemented on a decentralized control unit, e.g. the control unit 33 of the brake system 17, which also implements the brake pedal application in order to minimize the reaction times when adapting the brake pedal application. In this case, the control unit 33 of the brake system 17 receives all the sensor information from a drive control unit 35, the interior control unit 37 and driver assistance control units 39 via the data bus 19. according to FIG. 4, the driver assistance control unit 39 is connected to the vehicle surroundings sensor 9, while the interior control unit 37 is coupled to the driver monitoring sensors 7. The drive control unit 35 is connected to an accelerator pedal sensor and the control unit 33 of the brake system 17 is connected to the brake pedal sensor 3.This sensor data is processed in the control units 33, 35, 37, 39 in such a way that it can be used as features 21 afor predicting the brake pedal state by the model of the artificial intelligence 15 implemented in the control unit 33 of the brake system 17. The brake pedal state is calculated cyclically at a high frequency.Two examples of the adaptation of the brake pedal application by shifting the brake pedal characteristic curve, which represents the deceleration performance of the brake system over the brake pedal travel, as a function of the respectively detected brake pedal state are shown in FIG. 5. The result of the artificial intelligence model 15 classifies the brake pedal state into, for example, "brake pedal is free" and "brake pedal is blocked.". In the case of the unblocked brake pedal 3, the deceleration power can be set by the driver 23 on the basis of the existing brake pedal characteristic 41.If the brake pedal 3 is classified as locked, two locking scenarios can be predicted. In FIG. 5 a, a blocking position is detected, in which an object, for example a drinking bottle, a flip-flop sandale, a toy and the like, clamps the brake pedal 3 from behind.In this case, the model of the artificial intelligence predicts a brake pedal value of 80%, from which the brake pedal 3 cannot be pressed any further. In order nevertheless to achieve the maximum deceleration power (braking power), the brake pedal characteristic curve 41 that is present is adapted. Since the prediction is inaccurate and the object can move further, in this case the maximum deceleration power would be advanced to a brake pedal value of 50% (adapted brake pedal characteristic curve 43) and be spanned by the brake pedal travel thus shortened with a remaining dosability.In another blockage scenario that can be predicted by the artificial intelligence model 15, an object seems to prevent the brake pedal 3 from being fully released. This can occur, for example, by a foot mat resting on the brake pedal 3. In order that a slight deceleration power of the brake system is not required permanently and unintentionally, the minimum deceleration power is shifted to the predicted brake pedal value, which represents the contact of the foot mat on the brake pedal 3, to, for example, 20%. The brake pedal characteristic curve 43 is respanned from this point up to the maximum deceleration power.In order that the transition to the changed brake pedal application remains controllable for the driver 17, this change must take place abruptly (FIG. 5 a ) when the brake pedal is released or with a temporal ramp when the pedal is continuously actuated (FIG. 5 b ).As soon as the model of the artificial intelligence 15 has recognized a blocked brake pedal, the driver 23 is warned directly via a display device 47 (FIG. 1 ) and is requested to promptly approach a safe location and remove any object in the vicinity of the brake pedal 3. As additional measures, the other road users are informed about a possible hazardous situation via a Car2Car interface and the hazard warning system of the vehicle 1 is activated.After the object has been removed, the driver is requested via the display unit 47 to press the brake pedal 3 completely from 0% to 100% three times in succession. If the brake pedal is again released, the brake pedal application falls back to its basic state (brake pedal characteristic curve 41) again and the warning message is extinguished. As an additional safety measure, an entry in the fault memory is permanently stored in the brake control unit 33 so that, in the event of a later installation of the workshop, the expert can check the brake pedal 3 for damage.

Claims

System for controlling a braking behavior of a vehicle, comprising a central or decentralised control device (13, 33) which is connected to a multiplicity of vehicle sensors (5, 7, 9, 11), of which a sensor (5) for determining a position or movement of a brake pedal (3) is coupled to the latter, characterized in that the control device (13, 33) for detecting a blockage of the brake pedal (3) which is currently occurring or for predicting an imminent blockage has - a force sensor arranged on a brake pedal plate for checking the proportionality of the actuating force to the brake pedal travel, - an accelerator pedal travel sensor for detecting unintentional actuation, - an ultrasonic sensor arranged in the foot well of the vehicle (1) for detecting foreign bodies behind or next to the brake pedal (3) via distance detection, a camera arranged in the foot well of the vehicle (1) for detecting foreign objects behind or next to the brake pedal (3) by object identification on an image of the foot well, a force sensor arranged on the foot well floor of the vehicle (1) behind the brake pedal (3) for detecting foreign objects behind the brake pedal (3) by the brake pedal force transmitted from the latter to the foot well floor, an interior microphone arranged near the brake pedal (3) for receiving sounds which generate foreign objects which come into contact with the brake pedal (3), a seat mat or a force sensor in a seat cushion for determining a force displacement from the seat cushion to the backrest of the vehicle seat of the driver when the brake pedal (3) is actuated, a further interior camera for detecting a blocked brake pedal (3) on the basis of the facial expression of the driver, a further interior microphone for detecting the blocked brake pedal (3) by calling out the driver, a force sensor on the steering wheel for detecting the blocked brake pedal (3) by the driver clamping the steering wheel to a greater extent, a pulse sensor for detecting the blocked brake pedal (3) by a pulse increase of the driver, at least one environment detection sensor (11) for detecting the blocked brake pedal (3) by reducing the distance of the vehicle (1) from other vehicles or obstacles, wherein the control device (13, 33) has an artificial intelligence (15) which is trained once in a test vehicle on the basis of the sensor signals determined during a test run, wherein a predefined number of tuple (21) is generated, each of which comprises an explanatory feature (21a) formed by measurement data of the sensors (5, 7, 9, 11) and an explained feature (21b) in the form of the blocking state of the brake pedal (3) prevailing at the time of the measurement, wherein all generated tuple (21) are divided into a training set (25) and a test set (27), the amounts of which have no overlaps, with which a mapping function for predicting the blocking states of the brake pedal (3) is generated by means of one or more methods of monitored learning, wherein the prediction method of the model of the artificial intelligence (15) to be applied, which is implemented in the control device (13, 33) of a series vehicle (1), is determined on the basis of a degree of the deviation of a predicted unknown tuple (29) from the test set (27), in which the sensor data transmitted from the sensors (5, 7, 9, 11) to the control device (13, 33) are processed in such a way that they are used by the implemented model of the artificial intelligence (15) for predicting the brake pedal state for adapting the brake pedal tension for detecting the braking intention of the driver even in the case of a partial blocking of the brake pedal travel.Vehicle system according to Claim 1, characterized in that the control device of the test vehicle is designed, for training the model of the artificial intelligence (15), to generate training data using a simulation method which summarizes the physical relationships between the behavior of the driver, the behavior of the brake pedal (3) and the properties of the foreign body into a software function by following up a real driving maneuver of the test vehicle with the simulation and measuring data of virtual sensors.Vehicle system according to Claim 1 or 2, characterized in that a method of linear and polynomial regression and / or of the nearest neighbor classification and / or neural networks can be used for training the model of the artificial intelligence (15).Vehicle system according to Claim 1, 2 or 3, characterized in that the control unit (13, 33) of the series vehicle (1) is designed to execute a calculation of the brake pedal state cyclically at high frequency.Vehicle system according to at least one of the preceding claims, characterized in that the control unit (13, 33) classifies the brake pedal (3) and distinguishes between "brake pedal blocked" and "brake pedal unblocked" in the classification, wherein, if the brake pedal (3) has been classified as blocked, it is again distinguished whether the blockage is present at the end stop or at the zero position of the brake pedal (3).Vehicle system according to Claim 5, characterized in that a transition between different brake pedal applications takes place abruptly when the brake pedal (3) is released and in a time ramp when the brake pedal is permanently actuated.Vehicle system according to at least one of the preceding claims, characterized in that the control unit (13) is connected to a warning device (47) for outputting a warning when a critical state of the brake pedal (3) is determined.

Images