METHOD FOR CONTROLLING A VEHICLE BY A ASSISTANCE SYSTEM, ASSISTANCE SYSTEM AND VEHICLE

Description

[0001] The present invention relates to a method for controlling a vehicle by means of an assistance system, and in particular for autonomously moving a vehicle into a (variably) optimized dynamic driving state, and to a vehicle which is equipped with a system for controlling a vehicle by means of an assistance system and in particular for autonomously moving a vehicle into a (variably) optimized dynamic driving state.

[0002] Vehicles with autonomous driving functions are already known. In such vehicles, a control unit can handle both longitudinal control, i.e., accelerating and braking the vehicle, and lateral control, i.e., steering the vehicle. Vehicles with such autonomous driving functions can be semi-autonomous or fully autonomous. In semi-autonomous vehicles, the control unit only takes over vehicle control temporarily, for example, during certain driving situations. In fully autonomous vehicles, human intervention is no longer necessary. The degree of autonomy of a vehicle is divided into different levels, with vehicles with a higher degree of autonomy assigned to a higher level.

[0003] It can be assumed that in the future, there will be an increasing number of vehicles on the road that are fully autonomously controlled. Once a vehicle is classified as having an autonomy level of 4 or higher, the driving skills of any person inside the vehicle are no longer a relevant factor in controlling the vehicle or safely reaching a destination.

[0004] Even today, Level 2+ vehicles are capable of assessing situations and driving events, controlling the vehicle, assisting the driver, and / or warning the driver of hazards. Modern systems can often assess driving situations better than human drivers. This is made possible in particular by the enormous computing power of the hardware installed in the vehicle and the multitude of sensors available for monitoring the traffic situation. These sensors can often detect or identify signals that are imperceptible to humans. Examples of such signals include infrared, sonar, radar, lidar, and others. At least vehicles classified as Level 3 autonomous driving can take over control of the vehicle and its journey, at least for certain sections of the way.

[0005] The autonomy of vehicles offers new possibilities for designing interfaces for communication between humans and vehicles. It is expected that user interfaces will take on new forms in the future. Joysticks, touchpads, or even brain-computer interfaces (BCIs) in the more distant future could replace conventional control units such as steering and pedals. The entire interior of the vehicle could also be redesigned accordingly. In such vehicles, it will no longer be necessary for the driver to look ahead to observe the traffic situation.

[0006] One consequence of this increasing vehicle autonomy is that the interaction with the driver needs to be reconsidered. It is to be expected that with increasing vehicle autonomy, the driving experience for the occupants will diminish. A modified user interface may also mean that not all driver inputs are transmitted to the vehicle as before. For example, with joystick steering, direct transmission of the driver's input to the vehicle is no longer possible, as the input signals must be translated into signals for controlling the vehicle, such as its steering. Such modified input devices may also lead to reduced accuracy of the input signal. For instance, using a touchpad to input a driving signal could be expected to result in lower accuracy compared to a steering wheel. This could lead to less precise driving patterns and diminish the driving experience.

[0007] Even in vehicles with autonomy levels 2 and above, the vehicle can possess more knowledge and / or skills than a human driver. A more precise understanding of the driving function, the driving situation, the vehicle itself, and its capabilities in various driving situations allows such a vehicle to control itself in the best possible way. Particularly when combined with by-wire control of the actuators, it becomes possible to combine the driving function calculated by the vehicle with input from an occupant, such as the driver, and thus adapt the driving style to the driver's wishes, deviating from the calculated driving style.

[0008] Similarly, some users wish to be able to drive vehicles of autonomy levels 3, 4 or 5 themselves.

[0009] From US patent 2018 / 025 7 655 A1, a system and method for controlling the brakes and accelerator pedal of a vehicle are known. The aim is to predict a driver's intention to change the position of the brakes and accelerator pedal. Such a system is intended to reduce the reaction time until the brakes or accelerator pedal is fully applied by applying the appropriate pedal before the driver does. This action is predicted based on environmental data. By recognizing the environment, the system can predict when a driver intends to accelerate or brake. For example, this system can detect that the vehicle is approaching a ramp and prepare the vehicle in advance for the driver's anticipated acceleration, thus improving fuel efficiency through early acceleration.On the other hand, the system can detect that there is a danger and, in order to avoid an accident and increase the safety of the vehicle occupants, prevent the vehicle from, for example, colliding with a vehicle in front by preparing the braking process early.

[0010] A suitable parameter for determining the likely change in position of the brake or accelerator pedal has been identified as the acceleration of the accelerator pedal when its position is changed by the driver. For example, the immediate and complete return of the accelerator pedal to its original position can indicate an imminent application of the brake pedal. In other embodiments, the prediction is based on navigation data. For example, it is conceivable to predict impending acceleration after entering a highway using navigation data.

[0011] The system can adapt to specific drivers and their driving characteristics. A memory for driver-specific data is provided, and a driving profile assigned to each driver can be selected manually or through automated driver recognition. Other vehicle settings, such as seat position or mirror alignment, can also be automatically adjusted by corresponding vehicle actuators.

[0012] Patent application US 2007 / 0012499 A1 describes a driver assistance system for a vehicle that performs accelerator pedal actuation force control, drive force control, or brake force control in a manner appropriate to the traffic situation. To this end, the driver assistance system calculates a risk potential that indicates the degree of convergence between the vehicle and an obstacle located in front of the vehicle, based on a predefined control pattern. Based on this risk potential, the accelerator pedal actuation force, the vehicle's drive force, and / or the vehicle's brake force are then controlled. Furthermore, the driver's intention regarding acceleration and deceleration is also taken into account. For this purpose, the system proposes using the accelerator pedal's actuation state as a control variable.

[0013] In one example, a drive force control device adds or subtracts an additional drive force from the drive force requested by the driver to calculate a target drive force. Engine control is then performed based on this target drive force. The calculation of the additional drive force uses a parameter that correlates with the risk of a collision with a vehicle ahead. If the additional drive force is reduced from the drive force requested by the driver, the driver perceives this as the full drive force expected by the driver not being (or at least not completely) converted into acceleration, even though the accelerator pedal is depressed. If the modified (target) drive force is greater than the driving resistance, the driver experiences this as acceleration.If the revised (target) driving force is less than the driving resistance, this is perceived as negative acceleration or deceleration.

[0014] A system for assisted driving of a vehicle is known from US patent 2021 / 038 0 080 A1. This system measures the driver's reaction time and adjusts the control processes to account for or compensate for this delay. By anticipating a delayed input from the driver (for example, due to reaction time), the system can take proactive measures to avoid potentially dangerous situations. For instance, the vehicle can be braked early, giving the driver more time to react to a perceived hazard. The system can also intervene if the driver fails to react to a detected hazard within a predetermined timeframe. Furthermore, the system can take into account vehicle parameters such as the sensitivity of the steering wheel or brakes.

[0015] A brake signal generator is provided for control purposes, which can convert the control signals received from the vehicle computer or the electrical signals received from the pedal system into brake signals.

[0016] The vehicle equipped with this system can also include a sensor system that can provide data for an autonomous driving application. At least one sensor for detecting the driver's mental state is also described. This state can also be determined or predicted using artificial intelligence (AI). The AI ​​system is described as using cluster analysis in an unsupervised learning implementation to identify closely related data and / or employing a neural network to implement pattern recognition regarding the timing and nature of driver inputs within a given condition. This is then used to create a model for predicting the driver's level of attention based on the driver inputs.Furthermore, assistance systems for controlling a vehicle are known from WO 2010 / 101749 A1, US 2020 / 262478 A1 and US 2022 / 379883 A1.

[0017] Therefore, there is a need for a vehicle control system that can interact with the driver upon request and adapt the vehicle controls to the driver's wishes and / or abilities. Furthermore, there is a need for a method of controlling a vehicle using an assistance system that adapts the vehicle controls to the driver's wishes and / or abilities.

[0018] From a procedural standpoint, this problem is solved by a method according to claim 1.

[0019] This system allows the driver to intervene in the vehicle's controls and thus fulfill their wish to steer the vehicle. Furthermore, it allows the driver to influence the vehicle's behavior to a certain extent. At the same time, however, the system prevents the vehicle from entering an uncontrollable or critical state as a result of the driver's input. Even when the driver intervenes, the vehicle always remains on a route leading to the intended destination and maintains a stable condition at every point along that route.

[0020] It is preferably not necessary for the driver to independently control all functions for guiding the vehicle along the route to its destination. For example, it is conceivable that the driver only steers, while the (positive or negative) acceleration of the vehicle is handled by the assistance system. It would also be conceivable, for instance, that the driver could specify the shift points of a transmission (manually or semi-automatically) in order to operate the vehicle's engine, for example, in a particularly high and sporty rev range or in another, more energy-efficient rev range. Preferably, the driver can thus select which vehicle components he wishes to control independently and which controls should be performed by the assistance system.Preferably, the vehicle components not controlled by the driver are controlled by the assistance system, and in particular preferably, this control is adapted to vehicle control commands from the driver.

[0021] An assistance system that enables vehicle control as described above can be referred to as Advanced Driving Enhancement Systems (ADES), similar to the designation ADAS (Advanced Driver Assistance Systems), since it not only makes it possible to support the driver, but also to improve or enhance the driving experience.

[0022] Preferably, such a method allows the driver to directly influence the vehicle's driving behavior through their control commands. The driver preferably has the feeling that they are controlling the vehicle, or at least the vehicle components they have selected for self-control, and that only the other vehicle components are adapted to their inputs by the assistance system. This allows the driver's driving experience to be tailored to their preferences. The assistance system preferably adapts the control of other components not controlled by the driver to the driving style derived from the driver's control signals. Depending on the driver's and / or passenger's wishes, driving can thus be made more comfortable, efficient, and / or sportier. Preferably, this occurs without disrupting the driver's perceived sense of control.

[0023] Preferably, the calculation of the second driving route is performed taking into account data on a driving mode and / or driving characteristics of a driver, multiple drivers, and / or a group of drivers. This data can be derived from an immediate control command by the driver. However, it is conceivable and preferred that previous control commands are also used, for example, previous control commands during the current journey. Therefore, preferably, the driver's control commands are stored in a database. This database can contain entries for the current journey as well as for different journeys. Preferably, the database entries are assigned to specific drivers or vehicle occupants. This makes it possible, for example, after the assistance system detects a currently driving driver, to load a data record from the database that matches this driver (and, if applicable, other vehicle occupants).

[0024] The database may also include data records on driving modes that could be suitable for larger groups of drivers. Preferably, the database therefore includes one or more data records that correlate with a driving mode and / or driving characteristic selected from a group that includes a comfort driving mode, an energy-saving mode, a sport mode, a highway mode, a city driving mode, a long-distance mode, a work mode, a training mode, a passenger transport mode, a goods transport mode, a dangerous goods transport mode, and a synthetic data set, for example, from a simulation.

[0025] Preferably, a signal is transmitted electrically between the human-machine interface and an actuator for controlling the vehicle. It is particularly preferred that the vehicle, or at least some of its components, is controlled via X-by-Wire (for example, drive-by-wire, shift-by-wire, brake-by-wire, electronic accelerator pedal, and / or steer-by-wire). Preferably, a signal for controlling the vehicle is input at the human-machine interface. A human-machine interface can be, for example, a joystick, a touchpad, a touch-sensitive surface, a touchscreen, a rocker switch, a pedal, a rotary knob, and / or other devices.

[0026] This method allows the driver to take over certain controls, while the vehicle's assistance system handles the remaining controls to ensure safe vehicle operation. In particular, this method allows the assistance system to prevent the input signal from being immediately forwarded to an assigned actuator if it detects that a driver input would place the vehicle in a critical driving state. Instead, the signal is modified before transmission to the actuator so that the vehicle remains in a stable driving state when the actuator executes the action associated with the modified signal.

[0027] Through a process as described above, it is possible for the driver to experience improved vehicle handling compared to the driving style that would result from a direct implementation of their control signals. This improvement in vehicle handling is achieved without the driver having to adjust their own behavior and / or driving style. By selecting a specific driving mode, the type of improvement can be tailored to the desired objective. For example, such a process could improve driving comfort, energy efficiency, perceived sportiness, safety, and behavior in unfamiliar or overwhelming situations, as well as other situations, compared to the conditions that would arise from a direct implementation of the driver's control signals.

[0028] According to the invention, a human-machine interface comprises an optical and / or acoustic signal generator. This signal generator is designed as a display element that provides the driver with information about any deviation between the driver input expected by the assistance system and the input actually made. For example, the driver could be shown, on the one hand, a trajectory along which the vehicle would move if the actual steering command were given, and on the other hand, a target corridor within which the vehicle should move according to the assistance system's calculations. Acoustic and / or haptic signals could also indicate to the driver, for example, that the vehicle is leaving such a corridor.

[0029] With the increasing availability and use of (partially) autonomous vehicles, it is expected that the proportion of journeys on which a vehicle is (solely) driven by a human will decrease. Consequently, human driving experience will also decline. It is therefore to be expected that situations will occur more frequently in which a driver, through an unsuitable control signal, brings the vehicle they are driving (possibly with the support of an assistance system) into a critical driving situation. The procedure described above is suitable for avoiding such a critical driving situation. Nevertheless, it can be advantageous to provide the driver with feedback indicating that the control signal they entered was not implemented directly by the assistance system, but rather modified or even not implemented at all.Ideally, such feedback can trigger a learning effect in a driver, so that in the future such actions, which can lead to potentially critical driving situations, can be avoided.

[0030] Preferably, the driver is given feedback about any deviation between their driving instruction, as recorded by the human-machine interface, and an input expected by the assistance system or one of its components for the vehicle's movement along the pre-calculated path. This allows the driver to be alerted that their input had to be modified by the assistance system to keep the vehicle stable.

[0031] However, this procedure is not limited to avoiding critical driving situations. Rather, such a procedure can also provide the driver with feedback on other driving situations that are considered more favorable.

[0032] For example, depending on a driver's pre-selection, feedback could be provided if a more energy-efficient driving style would have been possible in a particular situation. Similarly, situation-dependent feedback could also be given regarding the possibility of a driving style with less lateral acceleration (e.g., in passenger transport) or a particularly safe driving style with increased safety margins (e.g., in passenger or hazardous goods transport). Accordingly, training for a pre-selected driving style is possible. Such a method could also be used in motorsport, for example, to demonstrate to a driver how to control the vehicle for faster cornering.

[0033] Preferably, data correlating with the feedback is stored in a database. A large number of such data points can be used, for example, to determine whether certain driver inputs, which could potentially lead to critical driving situations, are repeated. This data may allow conclusions to be drawn about the driver's preferred driving modes and / or driving characteristics. Furthermore, storing such data sets enables the recording and / or documentation of the driver's learning and / or training effect based on a large number of data points correlating with feedback. This data could be relevant, for example, to the insurance industry. In the motorsport application described above, a driver's training progress can be recorded, and training can be specifically tailored accordingly.

[0034] Preferably, the critical correlation threshold is set (by the driver or another authorized person or entity). This critical correlation threshold can represent a degree of decoupling between the driver's control command and that of the vehicle (or one or more of its actuators). Preferably, this critical correlation threshold can be set separately for individual control signal transmitter groups. By setting the critical correlation threshold in this way, it can, for example, be predetermined at what point the deviation between the driving instruction given by the driver and the predicted route of the vehicle and / or the predicted state of the vehicle is exceeded, at which point the immediate implementation of the driver's driving instruction no longer occurs.If, for example, the critical correlation threshold for accelerator pedal actuation (acceleration) is set lower than a critical correlation threshold for brake actuation, the system will intervene very early if a brake command is not received, whereas the system will only modify this input signal if there are very large deviations of an accelerator pedal actuation from an expected (and safe vehicle state) acceleration signal.

[0035] Preferably, unfavorable driving patterns and / or characteristics are identified based on data from a large number of a driver's journeys. The underlying data is preferably characteristic of at least one feature selected from a group that includes driver identification, vehicle occupant identification, vehicle identification, previous instances of the current driver falling below a critical correlation value, and the assignment of a previous instance of falling below a critical correlation value to a control signal or group of control signals. This makes it possible, for example, to frequently identify instances of a critical correlation value falling below a specific driver, which correlate with delayed braking and thus with unfavorable driving patterns and / or characteristics.The driver could be informed about this (for example, at regular intervals) to guide them towards improved driving habits. Furthermore, this data could be used to predict future driver inputs.

[0036] Preferably, at least one sensor monitors and / or detects the driver and / or a vehicle occupant. This makes it possible to assign a control signal precisely to that driver and / or vehicle occupant. Monitoring and identifying the occupants can be advantageous and / or necessary if an occupant takes over driving the vehicle and only intervenes in the vehicle control while driving. Furthermore, or additionally, the presence of certain vehicle occupants, such as babies or small children, could trigger a preselection of a different driving characteristic, for example, one with low (lateral) acceleration.It is also fortunate in this case that control signals from these vehicle occupants are not interpreted as control signals (or rather, any deviation of a driving instruction given by a baby or toddler from a driving instruction that correlates with the pre-calculated route of the vehicle and / or the pre-calculated vehicle state is interpreted as falling below the critical correlation threshold).

[0037] In a preferred variant of the procedure, the assistance system spans a virtual three-dimensional space whose dimensions These include a degree of decoupling of a driver input at the human-machine interface from a corresponding actuator signal, a degree of manipulation of the control signal given by the driver into a control of a corresponding actuator, and an intensity of feedback to the driver.

[0038] Preferably, the assistance system assigns each driver at least one point in this virtual space. This point correlates with a control and / or feedback behavior of the assistance system that is suitable for that driver and / or perceived as pleasant by that driver. The possibilities arising from such an assignment are illustrated in an exemplary procedure, particularly in connection with the Figures 5 and 6 However, the features shown are generally applicable within the scope of this invention.

[0039] Preferably, the point assigned to a driver is shifted along a preferably continuous curve in the three-dimensional space described above, depending on the driving situation. This makes it possible, for example, to provide the driver with a different driving feel during long-distance journeys than during leisurely driving (e.g., in city traffic) or slow driving (e.g., when maneuvering). It is therefore conceivable that each driver is assigned a curve in this three-dimensional space and that the aforementioned point, which preferably defines the correlation between the human-machine interface and an actuator controllable by this human-machine interface, is positioned on this curve depending on the driving situation.By systematically and jointly utilizing the aforementioned degrees of freedom, it is possible to provide the driver with a pleasant steering feel, which masks the (potentially significant) interventions of the assistance system. Nevertheless, the driver receives feedback on their actions and the desired (e.g., very direct) steering feel.

[0040] A point in this three-dimensional space, as described above, and / or a curve arranged within it, can be individually configured for each driver, determined by the assistance system (e.g., based on that driver's historical driving data), and / or preset by the manufacturer for a specific group of drivers. The driver may be able to modify the position of the point and / or the curve's shape. Preferably, the assistance system will recognize a driver's characteristics and adjust the curve accordingly. For example, if the driver's inputs correlate with a more sporty driving style, the degree of decoupling could be increased, and the driver could be simulated with enhanced feedback and a more direct / immediate intervention in the driving style, resulting in an even sportier driving experience than the vehicle actually employs.

[0041] Another example of a shift in the point or curve might occur when the assistance system is used to train a driver and a positive training effect is observed. In such a case, the degree of decoupling can be reduced, thus giving the driver more options for customizing the (actual) driving behavior of the vehicle.

[0042] Preferably, the method, i.e., at least one method step, is carried out using a machine learning model, in particular a trainable one. The model preferably comprises a set of parameters, in particular trainable ones, which are set to values ​​learned as a result of a training process.

[0043] At least one computer system is preferably connected to a data network, at least temporarily. This allows the computer system or computer system to receive updates and thus, for example, to be trained to recognize new hazards or to improve the accuracy of an algorithm for calculating a hazard risk (for example, resulting from falling below a critical correlation threshold). The data network can include a computer system with an artificial intelligence system that receives data from numerous computer systems (especially from different vehicles) and uses this data, along with data on actual and non-occurring hazard situations in individual vehicles, to generate improved algorithms for calculating hazard risks.

[0044] Preferably, the machine learning model is suitable for executing a (computer-implemented) computer procedure and determines in which (computer-implemented) perception and / or acquisition tasks are performed, for example, (computer-implemented) procedures for semantic segmentation and / or (computer-implemented) object classification. In object classification, a signal captured and / or represented in the measured values ​​(or in the sensor data characteristic of the measured values) is assigned to a (previously trained and / or predefined) class. The classes can be (among other things) a meaning (especially with regard to a driving instruction or a changed route) of a captured signal and / or a notification parameter characteristic of a meaning of the captured signal.

[0045] Preferably, the machine learning model is based on an (artificial) neural network (AI - Artificial Intelligence). Preferably, the sensor data (or data derived from it) are fed into the artificial neural network as input. Preferably, the artificial neural network maps the input variables to output variables as a function of a processing chain that can be parameterized (by the trainable or trained parameters).

[0046] Such a neural network can be designed, for example, as a deep neural network (DNN), in which the parameterizable processing chain has multiple processing layers, and / or as a convolutional neural network (CNN) and / or a recurrent neural network (RNN). Preferably, the parameterizable processing chain is parameterized during training. Preferably, datasets relating to the (base) datasets described above and / or training datasets are used as training data. Training preferably takes place using supervised learning. However, it would also be possible to train the artificial neural network using unsupervised learning, reinforcement learning, or stochastic learning.

[0047] The present invention is further directed to an assistance system according to claim 10.

[0048] Such an assistance system preferably has all the necessary components to carry out the procedure described above. In particular, it is designed, suitable, and / or intended to execute the procedure described above, as well as individual or all of the procedure steps already described above in connection with the procedure, either individually or in combination. Conversely, the procedure can be executed with all the features described within the assistance system, either individually or in combination.

[0049] As described above regarding the procedure, such an assistance system also makes it possible to guide the vehicle safely to the chosen destination even in the case of control signals from a driver that could potentially lead to a critical driving situation.

[0050] In the context of the present invention, a "second route" is understood to be a route that leads to the same destination as the first route. This destination preferably need not be the final destination of a journey, but may also be an intermediate destination. The second route need not necessarily differ from the first route in its geometric design. It would also be conceivable for it to be identical in shape to the first route. However, it preferably differs from the first route in at least one feature. This feature can be a characteristic parameter of the vehicle that is moving (or is intended to move) on this route. This characteristic parameter could, for example, be a speed, a distance to an obstacle, a steering angle, a different vehicle position and / or orientation, and / or other parameters.

[0051] Preferably, the human-machine interface of the assistance system includes a feedback device by which a result of the correlation value determination by the correlation value determination device can be transmitted to the driver. This makes it possible to inform the driver when their input at the human-machine interface deviates from an input that was expected at this human-machine interface for moving the pre-calculated driving path.

[0052] The feedback can also be output at a location different from where the control signal is input. This location could, for example, be a display device. This device could provide the driver with information about their input and, if it deviates from the expected input, about this deviation. It would also be conceivable to inform the driver in advance about an expected future input. For example, a view of a future section of the route could be displayed on the driver's screen. This section could be labeled with information about (driver) inputs expected at specific points along this section.If deviations from these expected inputs occur, the assistance system can infer the driver's intention to desire a different driving experience, and / or the assistance system could draw the driver's attention to this deviation in order to trigger a training effect and provoke improved driver inputs in the future.

[0053] Preferably, the strength of a feedback signal from the feedback device correlates with an amount by which the correlation limit is undershot. This allows the driver to know whether their input deviates significantly or only slightly from the expected input signal. Preferably, the driver can predefine a ratio between the strength of the feedback signal and the amount by which the signal falls below the expected value. By setting such a ratio, it is possible to ensure that a driver is not informed, not excessively informed, informed frequently, or always about deviations of their input from the expected input signal. Particularly for inexperienced drivers, selecting a suitable ratio can prevent information overload.

[0054] Preferably, the human-machine interface is electrically connected to an actuator for controlling the vehicle. This allows components of the assistance system to recognize, evaluate, and, if necessary, modify a signal transmitted by the human-machine interface via this connection, thereby corresponding to a change in the actuator's control signal. In a preferred embodiment, the human-machine interface is part of an X-by-Wire control system for the vehicle, in particular drive-by-wire, shift-by-wire, brake-by-wire, electronic accelerator pedal (e-accelerator pedal), and / or steer-by-wire.

[0055] Preferably, a data connection exists between the second computer unit and a database. The database preferably contains data on the driving characteristics of a driver, multiple drivers, a group of drivers, and / or different driving modes. This makes it possible to adapt the control signal generated or modified by the second computer unit to specific driving characteristics of a driver, multiple drivers, a group of drivers, and / or different driving modes. Preferably, at least one of the data records stored in the database correlates with a driving mode selected from a group that includes a comfort driving mode, an energy-saving mode, a sport mode, a highway mode, a city driving mode, a long-distance mode, a work mode, a training mode, a passenger transport mode, a goods transport mode, and a hazardous materials transport mode.For example, the driver or another authorized person or entity (such as the assistance system itself) can select a driving mode that suits the driver or that is likely to be closest to the desired driving profile.

[0056] Preferably, at least one, preferably several, and in particular preferably each of the devices from a group comprising the first computer device, the correlation value determination device, and the second computer device, is a computer-implemented device. In a particularly preferred embodiment, several of these devices are implemented on a common computer system. Such a system has proven to be particularly space-saving and, due to the shortened (data) transmission paths, particularly fast.

[0057] Preferably, the correlation value determination device and / or the second computer unit comprises an artificial intelligence system or is at least temporarily in data communication with such a system. The correlation value determination device and / or the second computer unit is preferably designed and configured to recognize a pattern in the deviation of the correlation between the driver's driving instruction and the expected driving instruction for following the predicted route. This pattern is preferably stored in a database, which forms the basis for future route calculations of the vehicle. Thus, control signals received from a driver can be used to recognize their desired driving style and / or driving characteristics, and based on this data, routes can be calculated in the future that more closely match the desired driving style and / or driving characteristics.

[0058] The present invention further relates to a vehicle, in particular a motor vehicle, comprising an assistance system as described above. The vehicle may in particular be a (motorized) road vehicle.

[0059] A vehicle can be a motor vehicle, which is in particular a semi-autonomous, autonomous (for example, at autonomy levels 2+, 3, 4, or 5 (according to the SAE J3016 standard)), or self-driving vehicle. Autonomy level 5 refers to fully automated vehicles. The vehicle can be controlled by a driver or drive autonomously. Furthermore, in addition to a road vehicle, the vehicle can also be an air taxi, an aircraft, or another means of transport or vehicle type, such as an air, water, or rail vehicle.

[0060] Further advantages and embodiments are shown in the attached drawings: These show: Fig. 1 a schematic representation of part of the proposed method according to one embodiment; Fig. 2 a schematic representation of part of the proposed method according to one embodiment; Fig. 3 an exemplary representation of the decoupling between the human-machine interface and the actuator; Fig. 4 another exemplary representation of the decoupling between the human-machine interface and the actuator; Fig. 5 a representation of a virtual space spanned by the possible degrees of freedom of decoupling between the human-machine interface and the actuator, as well as the feedback; Fig. 6 an exemplary arrangement of a feedback profile in the virtual space; Fig. 7 a schematic representation of the effect of the assistance system; Fig. 8 a schematic representation of selected driving modes; Fig. 9 another schematic representation of part of the proposed method according to one embodiment; Fig.Fig. 10 a schematic representation of a vehicle journey using a method variant; Fig. 11 another schematic representation of a vehicle journey using a method variant; and Fig. 12 another schematic representation of a vehicle journey using a method variant.

[0061] Fig. 1 A schematic representation of part of the proposed method according to one embodiment. The system 11 comprises a driver 2 designated by reference numeral 2, the assistance system 10 on which at least parts of the method are carried out, and the vehicle 1 and its actuators, which convert control signals from the assistance system 10.

[0062] Driver 2 performs an action marked with reference symbol A. This action correlates with a control signal u D which is transmitted to the assistance system 10. The control signal can be generated, for example, by a human-machine interface (not shown separately) based on an input (or action A) from driver 2.

[0063] Furthermore, the assistance system 10 receives information w Opt, which correlates with a desired driving experience 4 of the driver 2. This driving experience 4 is also referred to as the "optimal driving experience". The desired or optimal driving experience for the driver 2 can be communicated to the assistance system 10, for example, by the driver making a selection, or the assistance system 10 can independently select a presumably desired data set after driver identification, which correlates with the desired or optimal driving experience.

[0064] Based on the control signal u D and the information w Opt, the assistance system 10, which within the scope of this invention is also referred to as the "Driving Enhancing System" or "Advanced Driving Enhancement System", calculates a modified control signal u AS. This is a function g based on the input variables u D and w Opt, as well as further factors summarized under the symbol "x". These further factors include, in particular, the (possibly already calculated) route to a destination and environmental factors such as obstacles, other road users, road surface condition, route profile, and others.

[0065] The modified control signal u AS is then transmitted to vehicle 1 or its control unit and / or actuators. In step 6, the vehicle uses at least one, preferably several, suitable sensors (not shown) to determine information about the current driving state and / or driving situation. Other factors, such as the charge level of an energy storage device, the occupancy status of the passenger compartment, the load status of a cargo space, weather data, tire condition, and others, can also be included. The result of determining these factors (summarized here under the symbol "x" as described above) is transmitted to the assistance system and used there, as described above, as the basis for calculating the modified control signal u AS.

[0066] If the driver 2 wants feedback f, for example about a deviation of his given control signal u AS from the changed control signal u AS, this can be given to him via a corresponding feedback channel f.

[0067] In Fig. 2 A schematic representation of another part of the proposed method according to one embodiment is shown, with the focus on the steps performed by the assistance system 10. As in Fig. 1 The assistance system 10 receives the control signal u D, the information w Opt, and the data summarized under x, and calculates the modified control signal u AS based on this. The modified control signal u AS, and, if desired, the feedback signal f, are output to the vehicle or the driver as described above.

[0068] In one procedural variant, the steps designated with reference numbers 12 - 18 are provided, which are carried out by the assistance system 10.

[0069] In step 12, a prediction of the driver's action and / or an interpretation of the driver's situational intent is performed. For predicting the driver's action, data recorded under similar environmental conditions could be used, for example. If such a dataset exists, it can be used to determine how the driver has reacted in a similar situation in the past. There is a higher probability that the driver will act similarly in the now-detected situation.

[0070] Alternatively or additionally, the situational driver's wishes can be interpreted. This interpretation might, for example, conclude from the accelerator pedal being pressed that the driver wants to reach a (intermediate) destination particularly quickly or wants to drive in a particularly sporty manner.

[0071] In the optional step 14, the prediction may be improved to reflect the selected optimization goal. For example, if a comfortable ride is selected as the optimization goal, whereas the prediction was based on a sporty driving style, the prediction can be adjusted accordingly.

[0072] In step 16, the assistance system 10 determines a need for assistance to improve driver action A 2 in order to achieve the optimal driving experience 4. If, for example, increased comfort has been identified as the optimal driving experience 4, strong accelerations and / or sharp steering angles are not necessary to achieve this goal or are even counterproductive. In such a case, a need for assistance would be identified.

[0073] Furthermore, step 18 checks whether traffic and interaction safety is ensured. Driver 2's action 4 will not be executed if it would result in a vehicle state that endangers traffic safety. Likewise, driver 2's action 4 will not be executed if it is identified as unintended.

[0074] Fig. 3This diagram illustrates the potential decoupling between the human-machine interface and the actuator. Such decoupling is particularly possible with an X-by-Wire architecture. It offers numerous possibilities for driver assistance. It also allows the system to provide feedback to the driver if the assistance system deviates from a steering command. For example, the driver can be informed about steering torques. If a steer-by-wire system is used, the connection between the steering wheel and the front wheel and / or the accelerator and / or brake pedal can also be reduced to eliminate longitudinal forces and (partially) decouple the driver.

[0075] The level of driver assistance ranges from barely perceptible interventions to complete decoupling. This variation in the intervention possibilities between the human-machine interface and the actuator is referred to as "intervention dominance." This can be set between a minimum ("min") and a maximum ("max") value. A minimum value corresponds to manual driving M without intervention from an assistance system. Maximum intervention dominance exists when driver input E is completely decoupled from the action actually performed by the vehicle. Between these minimum and maximum positions lies a wide range in which an assistance system supports the driving or the driver; such a position is designated with the reference symbol G ("guided").

[0076] Various levels of decoupling or intervention dominance are conceivable. In the example shown, reference symbol I1 indicates a low level of driver assistance. This could be provided, for example, by audiovisual or acoustic information, such as a warning light or a signal tone.

[0077] Furthermore, the intervention dominance can be implemented differently in all support modalities and adapted to the situation. These modalities include, for example, longitudinal guidance, lateral guidance, visual information (through flashing lights, head-up display, or other means), and acoustic information. However, a further division of the interaction via the human-machine interface is particularly advantageous for the lateral guidance modality. In the following, these modalities will be simplified again for a summary analysis.

[0078] Further support could be offered, for example, at point I2. This could take place directly at the location where a corresponding reaction from the driver is expected. For example, it could be haptic feedback from the input device, such as a vibration of the steering wheel when there is a risk of leaving the designated lane.

[0079] I3 represents a further level of decoupling. With this level of decoupling, the vehicle independently takes control of certain functions. The driver no longer needs to control the input device associated with these functions. An example of this degree of decoupling is active vehicle guidance, such as lane-keeping steering torque (from a lane-keeping assist system). In such a case, however, the driver can always regain control of this function and, for example, force the vehicle to change lanes.

[0080] The situation is different with I4. In this range of decoupling levels, for example, the function of an emergency brake assist system is located. This system executes emergency braking independently of (at least some other) driver actions. For instance, when the emergency brake assist is activated, the driver is prevented from accelerating the vehicle. A corresponding input (e.g., at the accelerator pedal) would not be executed. Thus, at least partial decoupling is present.

[0081] Fig. 4 This shows another exemplary illustration of the decoupling between the human-machine interface and the actuator, using the example of varying the intervention dominance of the longitudinal dynamics control. At 0% intervention dominance M, the vehicle behaves analogously to the one in Fig. 3 The situation shown, with minimal intervention dominance M, is controlled exclusively manually.

[0082] Reference symbol E1 here indicates a low degree of decoupling, where, for example, an impending danger is indicated audiovisually or acoustically. However, the driver must react to such a warning, for example by manually applying the brakes.

[0083] A higher degree of decoupling is present at reference symbol E2. Here, some functions of the longitudinal dynamics control are decoupled and therefore no longer executable by the driver. For example, it is conceivable that a signal to accelerate the vehicle (such as pressing the accelerator pedal) is not converted into acceleration of the vehicle, or not fully.

[0084] In positions designated E3, E4, and E5, the longitudinal dynamics control is further decoupled by braking intervention from the assistance system. This braking intervention can vary in degree, starting with a slight braking intervention at E3, increasing through a progressive braking intervention at E4, and culminating in a complete braking intervention by the assistance system at E5.

[0085] In Fig. 5 This represents a virtual space defined by the possible degrees of freedom of decoupling between the human-machine interface and the actuator, as well as the feedback mechanism. Such a space is particularly well-suited to an X-by-Wire architecture and offers numerous possibilities for supporting the driver.

[0086] In this space, axis E spans a degree of decoupling, as found in the Figures 3 and 4This is illustrated. At the origin, there is no decoupling, and the driver steers manually. Complete decoupling is achieved at the end of axis E. There, the assistance system takes full control of the vehicle's steering, and any interventions by the driver on the steering wheel have no direct influence on the driving process. However, the ADES's interpretation of the driver's actions preferably remains intact, so the driver may still feel as if they have direct control over the driving process.

[0087] The axis extending between points FL and LS represents the degree of manipulation of the driver's control signal into the activation of a corresponding actuator. This axis represents the degree of freedom of an x-by-wire architecture, meaning that the feedback to the user does not have to directly correlate with the actual reactions acting on the vehicle. Through such manipulation of the control signal, a different actuator signal can be assigned to the same control signal depending on the driving situation.

[0088] At the origin L0, for example, the driver experiences no familiar steering feel when turning the steering wheel; instead, the steering wheel would rotate almost completely without torque into the driver's desired steering position. As the tuning moves towards the LS axis, a steering feel is simulated that correlates with the current driving situation, and the strength of the feedback can be varied, roughly comparable to adjusting power steering assistance. It would also be conceivable to modify the actuator signal depending on the vehicle speed to make maneuvering at low speeds easier for the driver. This modification could go so far that the driver perceives a fictitious steering feel FL, which bears no relation to the actual steering input applied at the actuator.

[0089] Another extreme is found at point LS, where the driver experiences a steering sensation as if every input were directly transmitted to the actuators and thus to the road. Since such direct transmission does not exist in a steer-by-wire system, it is conceivable that at point LS the driver is being simulated as if their inputs were being directly transmitted to the actuators.

[0090] The FF axis designates a direction in which the intensity of feedback to the driver increases. This feedback is independent of any manipulation of the control signal to an actuator signal. However, a feedback signal can superimpose itself on a control signal, and vice versa. For example, steering wheel vibration when approaching or crossing a lane marking is perceptible at the steering wheel and may superimpose itself on the force required to turn the steering wheel. This axis is comparable to the intervention options offered by systems without an X-by-Wire architecture, which lack the freedom of decoupling or steering feel simulation.

[0091] Fig. 6 shows an exemplary arrangement of a feedback profile in the virtual space, as it is in Fig. 5As illustrated by the curve shown (which is merely an example), such a system can occupy any point in the space spanned by the three degrees of freedom described. For example, based on a chosen correspondence between the control signal and the actuator signal on axis LS, feedback along axis FF can be selected for the driver. Feedback signals with a low steering torque (GLM) and a high steering torque (LM) are shown as examples. Such steering torque feedback to the driver can be provided even if the control signal is completely decoupled from the actuator signal. This means that even if the assistance system has taken over complete control of the steering and a steering wheel input would not translate into a change of direction, the driver can still receive feedback in the form of an actuator signal by transmitting a steering torque (LME) when the control signal is completely decoupled.

[0092] Fig. 7 Figure 10 schematically illustrates the effect of the assistance system 10. The assistance system 10 is capable of recognizing an action A performed by the driver 2. However, this action 2 of the driver does not necessarily have to be the best possible action A to achieve a goal desired by the driver 2. It is therefore merely an action that the driver subjectively considers suitable.

[0093] The assistance system also receives information from various sensors and about the selected destination. This "objective" information B, determined by or available to the system S, must also be taken into account.

[0094] The assistance system 10, illustrated here, must therefore understand the driver's subjective desire as accurately as possible and replicate it in such a way that the driver ideally cannot distinguish whether the driving conditions result from their direct actions or from the indirect replication of their subjective driving desire. Once the system is able to (at least partially) replicate this subjective driving desire, it becomes possible to improve the subjective driving desire in the direction of the "objective" driving goal B. A planning adaptation AB is therefore necessary. This shifts the driver's "subjective" desire towards an "objective" driving goal.

[0095] Fig. 8The diagram schematically shows a representation of selected driving modes FD1-FD3, which can also be an "objective" driving goal B or can be used to calculate an "objective" driving goal B. Driving modes such as Comfort FD1, Sporty FD2, and Ecological FD3 are shown by way of example only, using the graphics assigned to the reference symbols FD1-FD3.

[0096] In Fig. 9 Another schematic representation of a vehicle journey using a different procedure is shown. This procedure corresponds in part to the one in Fig. 2 The procedure described is such that only those process steps are described in detail that differ from the one in Fig. 2 The methods described differ.

[0097] Key difference compared to the one in Fig. 2The difference in the described method is that the input signal w Opt, which is used by the assistance system to calculate the modified control signal u AS, is not solely based on the driving behavior desired by the driver. Rather, according to this method variant, in addition to the desired driving experience 4 of the driver 2, the wishes and needs of other vehicle occupants 22, marked with reference numeral 24, are used to calculate the optimal driving experience 20 and thus also the input signal w Opt.

[0098] The information w Opt can thus be adjusted, preferably automatically, so that the journey is perceived as appropriate by most occupants. For example, if a baby is identified as a vehicle occupant, a journey with the lowest possible acceleration forces is considered particularly desirable, and the pre-calculated journey is optimized accordingly.

[0099] Fig. 10Figure 1 shows a schematic representation of a journey by vehicle 1 using a variant of the procedure. In the example shown, vehicle 1 is to travel along a curve 30 in direction R. The assistance system has calculated an initial path (not shown) for this purpose. However, in this example, the driver's steering signals do not correspond, or only partially correspond, to a journey along this initial path. For example, due to a lack of driving experience and / or the use of an unfamiliar human-machine interface such as a joystick or touchpad, the steering signals do not correspond to the calculated trajectory but to a meandering trajectory 32.

[0100] In this case, the assistance system recognizes the driver's intention to drive through the curve and combines this intention with a possible driving objective, namely to drive through the curve in a way that is gentle on the vehicle and / or with the lowest possible lateral acceleration peaks. In this case, the assistance system corrects the vehicle's course so that it continues along the second driving path, trajectory 34, despite the driver's contrary input. Thus, at least with regard to steering and at least temporarily, the driver is largely decoupled from the vehicle's control.

[0101] Fig. 11Figure 1 shows a further schematic representation of a vehicle 1 journey using a variant of the procedure. In the example shown, the driver is approaching an unfamiliar curve 30. Under the conditions under which he enters the curve, he would leave a consistency zone marked with reference numeral 38, in which this driver would normally be expected to drive, and the vehicle 1 would enter (danger) zone 45, at least temporarily. The assistance system recognizes this danger or adverse driving experience resulting from the driver's incorrect control input and corrects the driver's control signals so that the vehicle 1 travels safely or in a manner optimized for subjective perception along the newly calculated trajectory 34, which lies entirely within the safe target corridor 38 and the consistency zone of this driver.

[0102] The target corridor 38 could, for example, be predefined for the respective curve 30 by corresponding navigation data. Preferably, however, the target corridor 38 is calculated from the navigation data and a consistency range known to the assistance system for this driver. This allows information from previous curve passages by this driver to be used to calculate this curve passage. For example, varying driving skills and / or preferences of this driver can also be used to calculate a safe future curve passage. Fig. 12 is one of the Fig. 11A very similar situation involving the journey of vehicle 1 using a variant of the procedure is schematically depicted. However, the objective here is a high level of driving comfort. This objective cannot be achieved when negotiating curve 30 if vehicle 1 follows the driver's control command and enters the area marked with reference numeral 45 (for example, with high lateral acceleration). Therefore, the assistance system intervenes and calculates a second driving path 34 from the driver's control data, which lies entirely within the comfortable range 40. Subsequently, the vehicle is guided through curve 30 along trajectory 34 (possibly decoupled from driver commands by appropriately modifying the actuator control). Reference symbol list

[0103] 1 Vehicle 2 Driver 4 Desired, optimal driving experience 6 Current driving condition, current driving situation 10 Assistance system, Driving Enhancing System 11 System 12, 14, 16, 18 Procedure steps 20 Optimal driving experience 22 Vehicle occupants 24 Occupant wishes, needs 30 Road, curve 32 Trajectory, poor driving pattern 34 Calculated / driven trajectory 36, 37 Limits of the consistency range 38 Consistency range 40 Comfort range 41, 42 Limits of the comfort range 45 (Danger) range AA Action u D Control signal w Opt Information, u AS (optimized) control signal x Other factors F Feedback E Decoupling GA Assistance M Manual driving, manual control E1 - E5 Degree of decoupling I1 - I4 Degree of decoupling FF Feedback LSU Immediate steering feel FSF Fictitious steering feel L0 Origin GLM Low steering torque LM High steering torque LM Steering torque with complete decoupling S System B "Objective" information AB Planning adaptation FD1 - FD3 Driving modes R Direction

Claims

1. Method for controlling a vehicle (1) by means of an assistance system (10), the method comprising the steps of: a) a sensor device capturing surroundings data of a vehicle (1); b) calculating, by means of a first computer device, a first route for the vehicle (1) to a destination and a vehicle state at at least one point on this route on the basis of the surroundings data; c) at least one human-machine interface recording a driving instruction (A) from a driver (2); d) determining a correlation value on the basis of - the driving instruction (A) and - the precalculated route for the vehicle and / or the precalculated vehicle state; e) a second computer device checking whether the correlation value is below a critical correlation limit value and, if the correlation value is below the critical correlation limit value; converting a datum (uD), recorded by the human-machine interface and characteristic of the driving instruction (A) from the driver (2), into a control signal (uAS) which does not correspond to the recorded datum (uD) in a one-to-one relationship, wherein the control signal (uAS) controls the vehicle (1) or a vehicle component along a second route leading to the same destination, wherein the vehicle (1) adopts a stable vehicle state at every point on this second route, wherein the human-machine interface has an optical and / or acoustic signal generator, wherein the signal generator is a display element that shows the driver information relating to a discrepancy between the driver input expected by the assistance system and the input actually given.

2. Method according to claim 1, characterized in that the second route is calculated taking into account data, loaded from a database, relating to a driving mode and / or a driving characteristic of a driver (2), of a plurality of drivers, and / or of a group of drivers, wherein the driving modes (FD1 - FD3) are selected from a group which includes a comfort driving mode, an energy-saving mode, a sports mode, a highway mode, a city traffic mode, a long-distance mode, a work mode, a training mode, a passenger transport mode, a goods transport mode, and a dangerous goods transport mode.

3. Method according to either of the preceding claims, characterized in that a signal (uD) is electrically transferred between the human-machine interface and an actuator for controlling the vehicle (1), wherein the vehicle (1) is preferably controlled via X-by-wire control and an input of a signal (A) for controlling the vehicle (1) is given at the human-machine interface.

4. Method according to any of the preceding claims, characterized in that the driver (2) is given feedback (f) about a deviation of their state, recorded by the human-machine interface, and / or of their driving instruction (A), recorded by the human-machine interface, from an expected state and / or from an expected input for moving the vehicle (1) along the precalculated route, wherein a datum correlating to the feedback (f) is saved in a database in order to be able to record and / or document a learning and / or training effect on the driver on the basis of a plurality of data correlating to feedback.

5. Method according to any of the preceding claims, characterized in that the critical correlation limit value is set by the driver (2) or another authorized body, wherein this critical correlation limit value can be set separately for individual control signal generators and / or control signal generator groups.

6. Method according to any of the preceding claims, characterized in that unfavorable driving patterns and / or driving characteristics are detected on the basis of data from a plurality of journeys undertaken by a driver (2).

7. Method according to any of the preceding claims, characterized in that at least one sensor monitors and / or detects the driver (2) and / or a vehicle occupant (22).

8. Method according to any of the preceding claims, characterized in that at least one of the steps a) to f), preferably a plurality of the steps a) to f), further preferably a plurality of the steps b), d), e), and f), particularly preferably all of the steps b), d), e), and f), are performed with the aid of a computer, preferably as a computer-implemented method step.

9. Method according to any of the preceding claims, characterized in that a virtual, three-dimensional space is spanned by the assistance system, the dimensions of which space are - a degree to which a driver input at the human-machine interface is decoupled (E) from a corresponding actuator signal, - a degree to which the control signal given by the driver is manipulated (FL, LS) into control of a corresponding actuator, and - an intensity of feedback (FF) to the driver, wherein each driver is assigned at least one point in this virtual space by the assistance system, wherein this point correlates to a control and / or feedback behavior, suitable for that driver and / or perceived as pleasant by that driver, of the assistance system, wherein the point allocated to a driver is shifted along a preferably continuous curve in this three-dimensional space depending on the driving situation.

10. Assistance system (10) for at least partially autonomously controlling a vehicle (1) comprising a sensor device for capturing surroundings data, a first computer device for precalculating a first route for the vehicle (1) to a destination and a vehicle state at at least one point on this route on the basis of the surroundings data, at least one human-machine interface intended and configured to record a driving instruction (A) from a driver (2), a correlation value determination device intended and configured to determine a correlation value on the basis both of the driving instruction (A) and of the precalculated route for the vehicle and / or the precalculated vehicle state, and a second computer device intended and configured to check whether the correlation value is below a critical correlation limit value and, if the correlation value is below the critical correlation limit value, to convert a datum (uD), recorded by the human-machine interface and characteristic of the driving instruction (A) from the driver (2), into a control signal (uAS) which does not correspond to the recorded datum (uD) in a one-to-one relationship, wherein the control signal (uAS) correlates to a second route, leading to the same destination, on which the vehicle (1) can be guided in a stable vehicle state at every point, wherein the human-machine interface has an optical and / or acoustic signal generator, wherein the signal generator is a display element that shows the driver information relating to a discrepancy between the driver input expected by the assistance system and the input actually given.

11. Assistance system (10) according to claim 10, characterized in that the human-machine interface has a feedback device by means of which a result of the correlation value determination performed by the correlation value determination device can be transmitted to a driver (2), wherein a strength of a feedback signal (FF) from the feedback device correlates to an amount by which the correlation value is below the correlation limit value, wherein a ratio of the strength of the feedback signal (FF) to the amount by which the correlation value is below the correlation limit value can be specified by the driver.

12. Assistance system (10) according to claim 10 or 11, characterized in that the human-machine interface is electrically connected to an actuator for controlling the vehicle (1), wherein the human-machine interface is part of an X-by-wire controller of the vehicle (1).

13. Assistance system (10) according to any of claims 10-12, characterized in that the second computer device and a database are connected in terms of data, wherein in the data are stored data relating to the driving characteristics of a driver, of a plurality of drivers, a group of drivers, and / or relating to different driving modes, wherein at least one of the datasets stored in the database correlates to a driving mode (FD1 - FD3) selected from a group which includes a comfort driving mode, an energy-saving mode, a sports mode, a highway mode, a city traffic mode, a long-distance mode, a work mode, a training mode, a passenger transport mode, a goods transport mode, and a dangerous goods transport mode.

14. Assistance system (10) according to any of claims 9-13, characterized in that the correlation value determination device and / or the second computer device comprises an artificial intelligence system or is at least temporarily connected in terms of data to such a system, wherein the correlation value determination device and / or the second computer device is provided and configured to detect a pattern in the discrepancy of the correlation between the driving instruction from the driver and the expected driving instruction from the driver to follow the precalculations of a route and this pattern is preferably saved in a database which forms a basis for future calculations of a route for the vehicle.

15. Vehicle (1), in particular a motor vehicle, comprising an assistance system (10) according to any of claims 10-14.