Assistance system and assistance method for a vehicle

Abstract

The present disclosure relates to an assistance system (100) for a vehicle (10), comprising: - at least one processor module (110) which is designed to control vehicle functions and / or to request information from the vehicle functions on the basis of predetermined command inputs; - a user interface module (120) which is designed to receive verbal user inputs (BEN); and - an interface module (130) which is designed to communicate with the at least one processor module (110) and a large language model (LLM), wherein the interface module (130) is designed to: - provide the large language model (MML) with first configuration information (KFG1) relating to the predetermined command inputs for the vehicle functions; - provide the large language model (LLM) with the verbal user inputs (BEN) or information relating to the verbal user inputs (BEN); and - receive a conversion (KNV) of the verbal user inputs (BEN) from the large language model (LLM) based on the first configuration information (KFG1) and provide said conversion at least partially to the at least one processor module (110) in order to control vehicle functions accordingly and / or to request information from the vehicle functions.

Description

[0001] Assistance system and assistance procedure for a vehicle

[0002] The present disclosure relates to an assistance system for a vehicle, a vehicle having such an assistance system, an assistance method for a vehicle, and a storage medium for executing the assistance method. In particular, the present disclosure relates to the use of a large language model for interaction with a vehicle.

[0003] State of the art

[0004] In modern vehicles, various vehicle functions such as heating, entertainment, window lifts or navigation can already be operated via voice control. However, this voice control is developed specifically for the car and therefore only has limited performance. In particular, errors often occur in the interpretation of commands. The user cannot explain these errors to the system, as the system would not understand the explanation. This can lead to user dissatisfaction and cause them to not use voice control. Not using voice control can have a negative impact on road safety, for example in manually driven vehicles, as voice control has a lower potential for distraction from the driving task than using a touchscreen or similar control elements.

[0005] Disclosure of the invention

[0006] It is an object of the present disclosure to provide an assistance system for a vehicle, a vehicle with such an assistance system, an assistance method for a vehicle, and a storage medium for executing the assistance method, which enable error-reduced transmission of user instructions to a vehicle. In particular, it is an object of the present disclosure to provide improved voice control for a vehicle. This object is achieved by the subject matter of the independent claims. Advantageous embodiments are specified in the subclaims.

[0007] According to an independent aspect of the present disclosure, an assistance system for a vehicle, in particular a motor vehicle, is specified. The assistance system comprises at least one processor module configured to control vehicle functions and / or query information from the vehicle functions based on predetermined command inputs; a user interface module configured to receive voice user inputs; and an interface module configured to communicate with the at least one processor module and a large language model.

[0008] The interface module is configured to: provide the Large Language Model with first configuration information relating to the predetermined command inputs for the vehicle functions; provide the Large Language Model with the linguistic user inputs or information relating to the linguistic user inputs; and receive a conversion of the linguistic user inputs based on the first configuration information from the Large Language Model and provide it at least in part to the processor module in order to control vehicle functions accordingly and / or to retrieve information from the vehicle functions.

[0009] According to the invention, an interface is provided which mediates between a user and vehicle functions which can only process specific command inputs. For this purpose, the interface communicates with a large language model, which is provided, for example, in the form of a chatbot. The interface first communicates configuration information relating to the predetermined command inputs to the large language model. The interface then sends spoken user inputs (e.g., "the music is much too loud") to the large language model, which converts the spoken user inputs, based on the configuration information, into a format suitable for the vehicle functions (e.g., "reduce the radio volume"). Optionally, the large language model can also generate responses for the user based on the configuration information (e.g., "the volume has been adjusted").

[0010] As a result, improved voice control of vehicle functions can be enabled, as user instructions are transmitted to the vehicle with reduced errors. Furthermore, the interface according to the invention makes it possible to integrate new and / or different large language models at any time.

[0011] The user interface module and / or the interface module may comprise software components / algorithms configured to be executed on at least one processor and thereby to perform the functionalities of the respective module.

[0012] A large language model is a machine learning model from the field of generative artificial intelligence and is generally used to generate text-based content. Large language models are based on neural networks with a transformer architecture and use deep learning algorithms. An example of such a large language model is ChatGPT.

[0013] Preferably, the large language model is free of prior knowledge and acquires knowledge about the assistance system via the initial configuration information. In particular, the large language model can be developed and trained without prior knowledge of the rest of the assistance system and can only acquire knowledge about the assistance system via the initial configuration information.

[0014] Preferably, the large language model is configured for text-based communication. The interface module can be configured to provide the first configuration information relating to the predetermined command inputs in text form, and to provide the first linguistic user inputs or information relating to the linguistic user inputs in text form. Furthermore, the interface module can receive the conversion of the linguistic user inputs from the large language model into text form.

[0015] The vehicle functions preferably comprise or relate to one or more of the following functions: an entertainment system (e.g. radio); and / or an air conditioning system (e.g. heating, cooling, ventilation, seat heating, etc.); and / or window lifters, in particular electric window lifters; and / or navigation.

[0016] The at least one processor module can be configured to control at least one vehicle function based on the predetermined command inputs. For example, the at least one processor module can adjust a radio volume or open a window when a corresponding command is received.

[0017] The at least one processor module can be configured to query information from at least one vehicle function based on the predetermined command inputs. For example, the at least one processor module can query a current vehicle position when a corresponding command is received.

[0018] The term "predetermined command inputs," as used in the present disclosure, refers to a limited number of predefined expressions that can be processed / recognized by the processor module or vehicle function in order to be executed. In particular, the term "predetermined command inputs" refers to natural queries that can be processed / recognized by the vehicle or a speech recognition system implemented directly in the vehicle.

[0019] Preferably, the at least one processor module comprises, or is, a control unit. Control units are electronic modules that perform a variety of functions in vehicles and operate according to the so-called "EVA" principle: They receive signals from sensors and control elements, evaluate them, and control actuators that are responsible for converting the control unit's signals into a specific action. Control units are often networked via a data bus and can thus communicate with each other.

[0020] Preferably, the interface module is configured to transmit the first configuration information relating to the predetermined command inputs for the vehicle functions to the large language model during initialization of communication between the assistance system and the large language model. In particular, an interface operation can be configured using the configuration information before linguistic user inputs are transmitted to the large language model.

[0021] Preferably, the interface module is further configured to provide the large language model with second configuration information relating to a user response to be output by the user interface module relating to the control of the vehicle functions and / or the queried information; and to receive a user response based on the second configuration information from the large language model and to provide it at least partially to the user interface module for output to the user. This allows the user to be given feedback relating to the control of the vehicle functions and / or the queried information. For example, the user response "The navigation map is displayed" can be output in response to the "Navigation" command. In a further example, the user response "The radio is turned on" can be output in response to the "Radio" command.

[0022] Preferably, the interface module is configured to provide the first configuration information relating to the predetermined command inputs for the vehicle functions and the second configuration information relating to a user response to be output by the user interface module to the large language model substantially simultaneously or at the same time, for example during the initialization of the communication between the assistance system and the large language model.

[0023] The user interface module may comprise at least one input device for receiving the voice user input and optionally at least one output device for outputting the user response. The user interface module is preferably permanently installed in the vehicle.

[0024] The at least one input device may comprise a voice input device for receiving the voice user input and optionally a touch-sensitive input device, such as a touch pad, and / or a tactile input device, such as a switch (e.g., push switch and / or rotary switch) or other mechanically actuatable key elements.

[0025] The at least one output device may comprise at least one loudspeaker and / or at least one display device for outputting the user response. The at least one display device may comprise a display, in particular an LCD display, a plasma display, or an OLED display. Additionally or alternatively, the at least one display device may comprise a projection device configured to project information directly into the driver's field of vision, in particular onto a windshield.

[0026] Preferably, the interface module is configured to transmit a standard message to the Large Language Model if the received conversion of the linguistic user input does not correspond to any of the predetermined command inputs or cannot be processed / recognized by the processor module. This allows the Large Language Model to receive feedback upon receipt of an incorrect conversion.

[0027] Preferably, the standard message relates to the first configuration information relating to the predetermined command inputs and / or an explanatory user output to be output by the user interface module relating to the received conversion. If the Large Language Model sends a command that cannot be processed by the processor module or on-board computer, the processor module or on-board computer can respond with a standard message. The standard message can, for example, state that the command could not be processed and that this should be explained to the user. The Large Language Model can then generate an explanation that is output to the user by the user interface module (e.g., "The command was not understood. Please try again").

[0028] Preferably, the assistance system is configured to communicate with the Large Language Model according to a privacy setting. This allows the assistance system to decide which information is transmitted to the Large Language Model and which is not.

[0029] Preferably, the privacy setting can be specified or adjusted by the user. For example, the user can configure in the on-board computer which information should be made available to the Large Language Model to protect privacy. This configuration of the privacy setting can also be done when a command is used for the first time.

[0030] Preferably, the interface module is web-based. For example, the interface module can use a web URL to which the assistance system, such as the vehicle's onboard computer, can send requests and from which responses can be received. This type of integration allows for the rapid integration of any new large language model that appears on the market.

[0031] Preferably, the interface module is implemented as an application, in particular on a user's mobile device or a vehicle component (e.g., a control unit or a head unit). In other words, the user can install an app that provides the interface according to the invention. The term "mobile device" encompasses, in particular, smartphones, but also other mobile phones or cell phones, personal digital assistants (PDAs), tablets, notebooks, smart watches, smart glasses, and all current and future electronic devices equipped with communication technology.

[0032] Preferably, the Large Language Model is implemented in a central unit (e.g. server or backend) and / or cloud-based.

[0033] Preferably, the vehicle, in particular the assistance system, comprises a communication module. The vehicle's communication module can be configured for communication via a mobile network. The mobile network can be, for example, an LTE network or 5G network. This allows the vehicle to communicate with the Large Language Model (or the central unit and / or the cloud) via the mobile network in order to implement the functionalities according to the invention.

[0034] Preferably, the Large Language Model is a chatbot. A chatbot is a text-based dialogue system that allows communication with a technical system. The chatbot is configured for text input and output, enabling communication in natural language.

[0035] According to a further independent aspect of the present disclosure, a vehicle, in particular a motor vehicle, is specified. The vehicle includes the assistance system according to the embodiments of the present disclosure.

[0036] The term "vehicle" includes cars, trucks, vans, buses, mobile homes, motorcycles, etc., used to transport people, goods, etc. In particular, the term includes motor vehicles used to transport people.

[0037] According to a further independent aspect of the present disclosure, an assistance method for a vehicle, in particular a motor vehicle, is specified. The assistance method comprises transmitting, by an interface module, first configuration information relating to predetermined command inputs for vehicle functions to a large language model, wherein vehicle functions can be controlled and / or information from the vehicle functions can be queried based on the predetermined command inputs; receiving, by a user interface module, linguistic user inputs; transmitting, by the interface module, the linguistic user inputs or information relating to the linguistic user inputs to the large language model; receiving, by the interface module, a conversion of the linguistic user inputs from the large language model based on the first configuration information;and controlling at least one vehicle function and / or querying information from at least one vehicle function using the conversion of the linguistic user inputs received from the Large Language Model;

[0038] The assistance procedure can implement the aspects of the assistance system described in this document.

[0039] According to a further independent aspect of the present disclosure, a software (SW) program is provided. The SW program can be configured to run on one or more processors and thereby execute the assistance method described in this document.

[0040] According to a further independent aspect of the present disclosure, a storage medium is provided. The storage medium may comprise a software program configured to be executed on one or more processors and thereby to execute the assistance method described in this document.

[0041] According to a further independent aspect of the present disclosure, software with program code is specified. The software is configured to carry out the assistance method when the software runs on one or more software-controlled devices. According to a further independent aspect of the present disclosure, an assistance system for a vehicle is specified. The assistance system comprises one or more processors; and at least one memory connected to the one or more processors and containing instructions that can be executed by the one or more processors to carry out the assistance method described in this document.

[0042] According to a further independent aspect of the present disclosure, a chatbot system is provided. The chatbot system comprises one or more processors; and at least one memory connected to the one or more processors and containing instructions that can be executed by the one or more processors to receive (first) configuration information relating to predetermined command inputs for vehicle functions from an assistance system of a vehicle; to receive voice user inputs or information relating to the voice user inputs from the assistance system of the vehicle; and to generate a conversion of the voice user inputs based on the (first) configuration information and to transmit it to the assistance system of the vehicle.

[0043] A processor or processor module is a programmable computing unit, i.e. a machine or an electronic circuit that controls other elements according to given instructions and thereby drives an algorithm (process).

[0044] Short description of the drawings

[0045] Embodiments of the disclosure are illustrated in the figures and are described in more detail below. They show:

[0046] Figure 1 schematically shows a vehicle with an assistance system according to embodiments of the present disclosure, Figure 2 schematically shows an assistance system according to embodiments of the present disclosure, and

[0047] Figure 3 is a flowchart of an assistance method according to embodiments of the present disclosure.

[0048] Embodiments of the disclosure

[0049] In the following, unless otherwise stated, the same reference symbols are used for identical and equivalent elements.

[0050] Figure 1 schematically shows a vehicle 10 with an assistance system 100 according to embodiments of the present disclosure. Figure 2 schematically shows the assistance system 100 according to embodiments of the present disclosure.

[0051] The assistance system 100 comprises at least one processor module 110 configured to control vehicle functions and / or query information from the vehicle functions based on predetermined command inputs; and a user interface module 120 configured to receive voice user inputs from a vehicle user, in particular a driver.

[0052] The vehicle functions that can be controlled by the at least one processor module 120 using predetermined command inputs can, in some embodiments, include or relate to entertainment (e.g., radio), climate control (e.g., heating, cooling, ventilation, seat heating, etc.), power windows, and / or navigation, but are not limited thereto. For example, the at least one processor module 110 can adjust a radio volume or open a window when a corresponding command is received. In another example, the at least one processor module 110 can query a current vehicle position when a corresponding command is received. The term "predetermined command inputs," as used in the present disclosure, refers to a limited number of predefined expressions that can be processed / recognized by the processor module or the vehicle function in order to be executed.In particular, the term “predetermined command inputs” refers to natural requests that can be processed / recognized by the vehicle or by a speech recognition system implemented directly in the vehicle.

[0053] The user interface module 120 may include at least one input device 122 for receiving the voice user input from the vehicle user and optionally at least one output device 124 for outputting user responses. The user interface module 120 is preferably permanently installed in the vehicle 10.

[0054] The assistance system 100 further comprises an interface module 130, which is configured to communicate with the at least one processor module 110 and a Large Language Model (LLM), and optionally with the user interface module 120. The interface module 130 is provided, in particular, to communicate with the Large Language Model (LLM) with respect to the linguistic user inputs in order to generate processable command inputs for the at least one processor module 110. The Large Language Model (LLM) can be implemented in a central unit 20 (e.g., server or backend) and / or in a cloud-based manner.

[0055] The communication between the interface module 130, the at least one processor module 110 and the user interface module 120 can take place via a corresponding data bus of the vehicle 10.

[0056] Communication between the interface module 130 and the Large Language Model LLM can take place via a communication link 1, which uses, for example, a mobile network. The mobile network can be, for example, an LTE network or a 5G network. In some embodiments, the vehicle 10 can include a communication module with a SIM unit. The communication module can be configured to communicate with the central unit 20 via the mobile network. Thus, the vehicle 10 can communicate with the Large Language Model (or a central unit 20) via the mobile network in order to implement the functionalities according to the invention.

[0057] The interface module 130 is configured to communicate with the Large Language Model LLM regarding the linguistic user inputs in order to generate processable command inputs for the at least one processor module 110. For this purpose, the interface module 130 can be configured to: provide the Large Language Model LLM with first configuration information KFG1 regarding the predetermined command inputs for the vehicle functions; provide the Large Language Model LLM with the linguistic user inputs BEN or information regarding the linguistic user inputs BEN; and receive a conversion KNV of the linguistic user inputs BNE based on the first configuration information KFG1 from the Large Language Model LLM and at least partially provide it to the at least one processor module 110 in order to control vehicle functions accordingly and / or query information from the vehicle functions.

[0058] The interface module 130 thus mediates between an on-board computer, which can only process certain command inputs, and the user. In particular, the interface module 130 first communicates configuration information relating to the predetermined command inputs to the Large Language Model LLM. The interface module 130 then sends spoken user inputs (e.g., "the music is much too loud") to the Large Language Model LLM, which converts the spoken user inputs BEN into a format suitable for the vehicle functions (e.g., "reduce the radio volume") based on the configuration information KFG1. Optionally, the Large Language Model LLM can also generate responses for the user based on the configuration information (e.g., "the volume has been adjusted").In some embodiments, the interface module 130 is configured to transmit the first configuration information KFG1 relating to the predetermined command inputs for the vehicle functions to the Large Language Model LLM during an initialization of communication between the assistance system 100 and the Large Language Model LLM. In particular, an interface operation can be configured using the configuration information before linguistic user inputs BEN are transmitted to the Large Language Model LLM.

[0059] Optionally, the interface module 130 can be further configured to provide the Large Language Model LLM with second configuration information KFG2 relating to a user response to be output by the user interface module 120 relating to the control of the vehicle functions and / or the queried information; and to receive a user response based on the second configuration information KFG2 from the Large Language Model LLM and to provide it at least partially to the user interface module 120 for output to the user. This allows the user to be given feedback relating to the control of the vehicle functions and / or the queried information. For example, the user response "The navigation map is displayed" can be output in response to the "Navigation" command. In another example, the user response "The radio is turned on" can be output in response to the "Radio" command.

[0060] The interface module 130 can be configured to provide the first configuration information KFG1 relating to the predetermined command inputs for the vehicle functions and the second configuration information KFG2 relating to a user response to be output by the user interface module 120 to the Large Language Model LLM essentially simultaneously or at the same time, for example, during the initialization of the communication between the assistance system 100 and the Large Language Model LLM. Preferably, the Large Language Model LLM is a chatbot. A chatbot is a text-based dialog system that allows communication with a technical system. The chatbot is configured for text input and output, so that communication in natural language is possible.

[0061] Example first configuration information KFG1 and second configuration information KFG2 can be output in text form to the Large Language Model LLM and are as follows:

[0062] Now I want you to act as an interface for a vehicle's onboard computer. If I begin a message with #, consider it a user request and respond in the format shown in the next two lines:

[0063] Command: “Vehicle Command”

[0064] Answer: “User response”

[0065] Vehicle command is a command to the vehicle, and user response is the text to be displayed to the user. The vehicle supports the commands "Navigation" and "Radio." Navigation displays the map, while Radio turns on the radio.

[0066] This configuration can also be flexibly adapted later by the interface module 130 transmitting further configuration information KFG1, KFG2 to the Large Language Model LLM, such as:

[0067] The vehicle’s on-board computer now also supports the “Radio off” command, which turns off the radio.

[0068] If the user interface module 120 now receives a spoken user input "#Turn on the radio," this is transmitted by the interface module 130 to the Language Model LLM. The Language Model LLM returns a corresponding conversion KNV, ie, a predetermined command input and a user response to be output: Vehicle command: Radio

[0069] User response: Radio is turned on

[0070] In some embodiments, the interface module 130 can be configured to transmit a standard message to the Large Language Model LLM if the received conversion KNV of the linguistic user inputs does not correspond to any of the predetermined command inputs BEN or cannot be processed / recognized by the processor module 110. This allows the Large Language Model LLM to receive feedback upon receipt of an erroneous conversion.

[0071] The standard message can include the first configuration information KFG1 relating to the predetermined command inputs and / or an explanatory user output to be output by the user interface module 120 relating to the received conversion KNV. If the Large Language Model LLM sends a command that cannot be processed by the processor module 110 or on-board computer, the processor module 110 or on-board computer can respond with a standard message. The standard message can, for example, state that the command could not be processed and that this should be explained to the user. The Large Language Model LLM can then generate an explanation that is output to the user by the user interface module 120 (e.g., "The command was not understood. Please try again").

[0072] In some embodiments, the assistance system 100 is configured to communicate with the Large Language Model LLM according to a data protection setting. This allows the assistance system 100 to decide which information is transmitted to the Large Language Model LLM and which is not. The data protection setting can be specified or set by the user. For example, the user can configure in the on-board computer which information should be made available to the Large Language Model LLM in order to protect privacy. This configuration of the data protection setting can also occur when a command is used for the first time. In some embodiments, the interface module 130 is web-based. For example, the interface module 130 can use a web URL to which the assistance system 100, such as an on-board computer of the vehicle 10, can send requests and from which responses can be received.This type of integration makes it possible to quickly integrate any new large language model that appears on the market.

[0073] In further embodiments, the interface module 130 is implemented as an application, in particular on a user's mobile device or a vehicle component (e.g., a control unit or a head unit). In other words, the user can install an app that provides the interface according to the invention.

[0074] Figure 3 schematically shows a flowchart of an assistance method 300 according to embodiments of the present disclosure. The assistance method 300 can be implemented by appropriate software executable by processors (e.g., a CPU).

[0075] The assistance method 300 comprises, in block 310, transmitting, by an interface module, first configuration information relating to predetermined command inputs for vehicle functions to a large language model, wherein vehicle functions can be controlled based on the predetermined command inputs and / or information from the vehicle functions can be queried; in block 320, receiving, by a user interface module, linguistic user inputs; in block 330, transmitting, by the interface module, the linguistic user inputs or information relating to the linguistic user inputs to the large language model; in block 340, receiving, by the interface module,a conversion of the linguistic user inputs from the Large Language Model based on the first configuration information; and in block 350, a control of at least one vehicle function and / or a query of information from at least one vehicle function using the conversion of the linguistic user inputs received from the Large Language Model. According to the invention, an interface is provided that mediates between a user and vehicle functions that can only process certain command inputs. For this purpose, the interface communicates with a Large Language Model, which is provided, for example, in the form of a chatbot. The interface first communicates configuration information relating to the predetermined command inputs to the Large Language Model. The interface then sends linguistic user inputs (e.g., "the music is much too loud") to the Large Language Model.which converts the spoken user input based on the configuration information into a format suitable for the vehicle's functions (e.g., "reduce radio volume"). Optionally, the Large Language Model can also generate responses for the user based on the configuration information (e.g., "the volume has been adjusted").

[0076] As a result, improved voice control of vehicle functions can be enabled, as user instructions are transmitted to the vehicle with reduced errors. Furthermore, the interface according to the invention makes it possible to integrate new and / or different large language models at any time.

[0077] Although the invention has been illustrated and explained in detail by means of preferred embodiments, the invention is not limited by the disclosed examples, and other variations may be derived therefrom by those skilled in the art without departing from the scope of the invention. It is therefore clear that a multitude of possible variations exist. It is also clear that the embodiments mentioned by way of example are truly only examples and should not be construed as limiting the scope, possible applications, or configuration of the invention in any way.Rather, the preceding description and the description of the figures enable the person skilled in the art to implement the exemplary embodiments in concrete terms, whereby the person skilled in the art, with knowledge of the disclosed invention concept, can make various changes, for example with regard to the function or the arrangement of individual elements mentioned in an exemplary embodiment, without departing from the scope of protection defined by the claims and their legal equivalents, such as further explanations in the description.

Claims

Patent claims 1. An assistance system (100) for a vehicle (10), comprising: at least one processor module (110) configured to control vehicle functions based on predetermined command inputs and / or to query information from the vehicle functions; a user interface module (120) configured to receive voice user inputs (BEN); and an interface module (130) configured to communicate with the at least one processor module (110) and a large language model (LLM), wherein the interface module (130) is configured to: provide the large language model (MML) with first configuration information (KFG1) relating to the predetermined command inputs for the vehicle functions; provide the large language model (LLM) with the voice user inputs (BEN) or information relating to the voice user inputs (BEN);and to receive a conversion (KNV) of the linguistic user inputs (BEN) from the Large Language Model (LLM) based on the first configuration information (KFG1) and to provide it at least partially to the at least one processor module (110) in order to control vehicle functions accordingly and / or to query information from the vehicle functions.; 2. Assistance system (100) according to claim 1, wherein the interface module (130) is further configured to: provide the Large Language Model (LLM) with second configuration information (KFG2) relating to a user response to be output by the user interface module (120) relating to the control of the vehicle functions and / or the queried information; and to receive a user response based on the second configuration information (KFG2) from the Large Language Model (LLM) and to provide it at least in part to the user interface module (120) for output to the user.

3. Assistance system (100) according to claim 1 or 2, wherein the interface module (130) is configured to transmit a standard message to the Large Language Model (LLM) if the received conversion (KNV) of the linguistic user inputs (BEN) does not correspond to any of the predetermined command inputs, wherein the standard message relates to the first configuration information (KFG1) relating to the predetermined command inputs and / or an explanatory user output to be output by the user interface module (120) relating to the received conversion (KNV).

4. Assistance system (100) according to one of claims 1 to 3, wherein the assistance system (100) is configured to communicate with the Large Language Model (LLM) according to a data protection setting, in particular wherein the data protection setting can be specified by the user.

5. Assistance system (100) according to one of claims 1 to 4, wherein the interface module (130) is web-based, or wherein the interface module (130) is implemented as an application, in particular on a mobile terminal of the user or a vehicle component.

6. Assistance system (100) according to one of claims 1 to 5, wherein the Large Language Model (LLM) is implemented in a central unit (20) and / or cloud-based.

7. Assistance system (100) according to one of claims 1 to 6, wherein the large language model (LLM) is free of prior knowledge and obtains knowledge about the assistance system via the first configuration information (KFG1).

8. Vehicle (10), in particular a motor vehicle, comprising the assistance system (100) according to one of claims 1 to 7.

9. Assistance method (300) for a vehicle (10), comprising: Transmitting (310), by an interface module (130), first configuration information (KFG1) relating to predetermined command inputs for vehicle functions to a Large Language Model (LLM), wherein vehicle functions can be controlled based on the predetermined command inputs and / or information from the vehicle functions can be queried; Receiving (320), by a user interface module (120), linguistic user inputs (BNE); Transmitting (330), by the interface module (130), the linguistic user inputs (BEN) or information relating to the linguistic user inputs (BEN) to the Large Language Model (LLM); Receiving (340), by the interface module (130), a conversion (KNV) of the linguistic user inputs (BNE) from the Large Language Model (LLM) based on the first configuration information (KFG1); and Controlling (350) at least one vehicle function and / or querying information from at least one vehicle function using the conversion (KNV) of the linguistic user inputs (BEN) received from the Large Language Model (LLM).

10. A storage medium comprising a software program configured to be executed on one or more processors and thereby to execute the assistance method (300) according to claim 9.

11. A chatbot system (20) comprising one or more processors and at least one memory connected to the one or more processors and containing instructions executable by the one or more processors to: to receive configuration information (KFG1) relating to predetermined command inputs for vehicle functions from an assistance system (100) of a vehicle (10); to receive voice user inputs (BNE) or information relating to the voice user inputs (BNE) from the assistance system (100) of the vehicle (10); and to generate a conversion (KNV) of the voice user inputs (BNE) based on the configuration information (KFG1) and to transmit it to the assistance system (100) of the vehicle (10).

Images