Computer-implemented method for controlling at least one vehicle function of a vehicle and vehicle

Abstract

The invention relates to a computer-implemented method (100) for controlling at least one vehicle function (18) of a vehicle (10), wherein the vehicle (10) has a first sensor device (14) for acquiring first measurement data indicating a driver's health status and a second sensor device (16) for acquiring second measurement data indicating a driving situation of the vehicle (10) while driving, comprising at least the following steps: determining (102) a driver's health status from received first measurement data using a fatigue agent, a stress agent, and / or an agent for detecting potentially medical abnormalities; determining (104) a driving situation from received second measurement data using a situation-context agent; checking (106) whether the determined health status is a critical health condition for driving the vehicle (10);In the event of a critical health condition: Generate (108) at least one control signal for at least one vehicle function (18) using an intervention agent. The procedure (100) can be used to proactively increase driving safety.

Description

[0001] The invention relates to a computer-implemented method for controlling at least one vehicle function of a vehicle and a vehicle.

[0002] Modern vehicles are increasingly integrating sensors to monitor technical vehicle conditions. At the same time, there is growing interest in recording the driver's physiological and cognitive states to improve safety and comfort. However, existing systems typically only detect individual parameters (e.g., fatigue detection via cameras) and react with simple warnings.

[0003] Existing systems are limited by: Reactive logic: Classical fatigue warning systems often only react to clearly visible signs; Siloed data usage: Physiological, vehicle, and environmental data are not evaluated in an integrated manner; Lack of continuous health monitoring: Mechanisms for detecting stress, fatigue, or potential health problems are missing; No adaptive feedback: Vehicle functions are not dynamically adjusted to the driver's condition; Lack of preventative mechanisms: Recommendations for breaks, breathing exercises, or relaxation modes are not based on real-time analysis. This limits both road safety and the driver's physical well-being.

[0004] The object of the invention is to provide a computer-implemented method and a vehicle that proactively increase driving safety.

[0005] The problem is solved by the features of the independent claims. Advantageous further developments are the subject of the dependent claims and the following description.

[0006] According to a first aspect, a computer-implemented method for controlling at least one vehicle function of a vehicle is described, wherein the vehicle has a first sensor device for acquiring first measurement data indicating a driver's health status and a second sensor device for acquiring second measurement data indicating the vehicle's driving situation, comprising at least the following steps: determining the driver's health status from received first measurement data using a fatigue agent, a stress agent, and / or an agent for detecting potentially medical abnormalities; determining the driving situation from received second measurement data using a situation-context agent; and checking whether the determined health status is a critical health condition for driving the vehicle.In the event of a critical health condition: Generate at least one control signal for at least one vehicle function using an intervention agent.

[0007] The present invention describes a large-language model (LLM) and AI-supported multi-agent system that continuously monitors the driver's health and stress levels and dynamically links this information to the current driving situation. In principle, any of the agents mentioned above can be configured as an LLM. In particular, the intervention agent can be provided as an LLM. An agent is understood to be a computer program that is capable of independent and self-directed behavior to a certain extent.

[0008] The system uses multimodal sensors (camera-, heart rate- and motion-based) as well as situational vehicle data to: detect fatigue, stress, overexertion or potential medical events, control adaptive vehicle functions (e.g. seat position, climate control, assistance systems), initiate proactive safety interventions, and provide health-promoting recommendations (breaks, breathing exercises, driving mode changes).

[0009] This method enables biometric and situational fusion, combining physiological data, vehicle data, navigation context, and environmental influences in a unique way. Furthermore, a multi-agent architecture can be implemented: specialized agents can monitor various driver states, interpret driving scenarios, and coordinate actions. At least one proactive safety intervention can also be provided: the system can detect emerging risks before critical situations arise. In addition, health prevention can be provided, not just health safety, focusing on stress reduction, cognitive relief, and long-term health. Finally, adaptive vehicle interaction can be provided, allowing for automatic adjustments to interior parameters, driver assistance levels, warning intensities, and route guidance.Furthermore, LLM-based communication can be provided: This allows for natural language interaction to explain recommendations and clarify the situation.

[0010] The fatigue agent can be trained to detect tiredness, microsleep, and slowed reactions. The stress agent can be trained to analyze overload, increased tension, or situational uncertainty. The agent for detecting potentially medical abnormalities can be trained to identify potential medical abnormalities (e.g., changes in heart rhythm). The situation-context agent can be trained to assess the current driving situation (weather, traffic jam, complexity, hazard). The intervention agent (control instance) can be trained to decide on appropriate measures, particularly from the group of warnings, interior adjustments (lights, climate, seat), activation of driver assistance systems, and / or recommendations or mandatory breaks (stage-based).

[0011] At least one of the agents mentioned above can use a deep learning model for biometric pattern recognition. Furthermore, at least one of the agents mentioned above can have a learning management model (LLM) for interpretation, context analysis, and / or natural communication. Multimodal fusion can be provided via transformer models.

[0012] The control signal can, for example, control the following vehicle functions: climate and seat adjustment, ambient lighting, music or soundscapes and / or increase autonomous driving level / activate assistance mode.

[0013] The system is designed to jointly interpret biometric and vehicle dynamics data, operate via a networked multi-agent system, react proactively and mitigate health risks, control adaptive vehicle functions, enable natural communication through LLMs, and, for the first time, combine safety and health monitoring in an integrated, intelligent architecture. This invention thus opens up new possibilities for safety, well-being, and human-centered mobility.

[0014] According to some embodiments, it is conceivable that the first measurement data could include camera data from an interior camera, heart rate data from a seat and / or steering wheel sensor and / or skin conductance data from a sensor to detect the skin conductance of the driver.

[0015] According to some embodiments, it is conceivable that the second measurement data could include steering wheel and / or pedal movement data and / or vehicle dynamics data.

[0016] According to some embodiments, it is conceivable that the situation-context agent could further use navigation and / or traffic data, which in particular indicate congestion, weather conditions and / or the complexity of the route, to determine the driving situation.

[0017] According to some embodiments, it is conceivable that the fatigue agent could be designed to detect fatigue, microsleep and / or slowed reactions as a health condition.

[0018] According to some examples, it is conceivable that the stress agent could be trained to detect overload, increased tension and / or situational uncertainty.

[0019] According to some embodiments, it is conceivable that the intervention agent could be designed to provide at least one warning signal, one control signal for an interior vehicle function, one start signal for at least one assistance system and / or to control the vehicle to drive to the nearest parking space.

[0020] According to a second aspect, a computer program product is described, comprising instructions that, when the program is executed by a computer, cause it to perform the steps of the procedure according to the preceding description.

[0021] The advantages, effects, and further developments of the computer program product result from the advantages, effects, and further developments of the method described above. Therefore, reference is made to the preceding description in this regard. A computer program product can be understood, for example, as a data carrier on which a computer program element is stored, containing instructions executable by a computer. Alternatively or additionally, a computer program product can also be understood, for example, as a permanent or volatile data storage medium, such as flash memory or main memory, that contains the computer program element. However, this does not exclude other types of data storage media that contain the computer program element.

[0022] According to a third aspect, a vehicle is described comprising at least one vehicle function, a first sensor device for acquiring first measurement data indicating a health value of a driver, and a second sensor device for acquiring second measurement data indicating a driving situation of the vehicle while driving, and comprising at least one computer designed to receive the first measurement data, the second measurement data and to execute the steps of the procedure according to the preceding description.

[0023] The advantages, effects, and further developments of the vehicle result from the advantages, effects, and further developments of the procedure described above. To avoid repetition, reference is therefore made to the preceding description in this regard.

[0024] The invention is described below with reference to an exemplary embodiment and the accompanying drawing. The drawing shows: Fig. 1. A flowchart of the procedure; and Fig. 2 a schematic representation of the vehicle.

[0025] The computer-implemented method for controlling at least one vehicle function is described in accordance with Fig. 1 hereinafter referred to in its entirety by reference numeral 100.

[0026] Procedure 100 can be used in conjunction with a Fig. The procedure is carried out using the vehicle 10 shown in Figure 2. The vehicle 10 has at least one vehicle function 18, a first sensor device 14 for acquiring initial measurement data, and a second sensor device 16 for acquiring secondary measurement data. The initial measurement data can indicate a driver's health status. The secondary measurement data can indicate the vehicle 10's driving situation. Furthermore, the vehicle 10 has at least one computer 12 configured to execute the steps of procedure 100.

[0027] The computer 10 can be connected to the first sensor device 14, the second sensor device 16 and the at least one driving function 18 via wireless or wired signal connections 20, 22, 24.

[0028] Procedure 100 shows according to Fig.Step 102 is performed, in which the driver's health status is determined from received initial measurement data using a fatigue agent, a stress agent, and / or an agent for detecting potentially medical abnormalities. The initial measurement data may include camera data from an interior camera, heart rate data from a seat and / or steering wheel sensor, and / or skin conductance data from a sensor to measure the driver's skin conductance.

[0029] The fatigue agent can be trained to detect fatigue, microsleep and / or slowed reactions as a health condition.

[0030] The stress agent can be trained to recognize overload, increased tension and / or situational uncertainty.

[0031] According to step 104, a driving situation is determined from received secondary measurement data using a situation-context agent. The secondary measurement data can include steering wheel and / or pedal movement data and / or vehicle dynamics data.

[0032] The situation context agent can further use navigation and / or traffic data, which in particular indicate congestion, weather conditions and / or the complexity of the route, to determine the driving situation.

[0033] According to step 106, it is checked whether the determined health condition is a critical health condition for driving the vehicle 10.

[0034] If a critical health condition is present, step 108 is performed, in which at least one control signal for at least one vehicle function 18 is generated by means of an intervention agent.

[0035] The intervention agent may be trained to provide at least one warning signal, one control signal for an interior vehicle function 18, one start signal for at least one assistance system and / or to steer the vehicle 10 to the nearest parking space.

[0036] The example described above does not in any way limit the invention. Rather, the invention can be modified in many different ways.

[0037] All of the features of the invention described above can be essential to the invention, either alone or in combination with one another.

Claims

[1] Computer-implemented method (100) for controlling at least one vehicle function (18) of a vehicle (10), wherein the vehicle (10) has a first sensor device (14) for acquiring first measurement data indicating a health value of a driver and a second sensor device (16) for acquiring second measurement data indicating a driving situation of the vehicle (10) while driving, comprising at least the following steps: a. Determining (102) the health status of the driver from initial measurement data received using a fatigue agent, a stress agent and / or an agent for detecting potentially medical abnormalities; b. Determining (104) a driving situation from received second measurement data using a situation-context agent; c. Check (106) whether the identified health condition is a critical health condition for driving the vehicle (10); d. If a critical health condition is present: Generate (108) at least one control signal for at least one vehicle function (18) by means of an intervention agent. [2] Computer-implemented method (100) according to claim 1, characterized by , that the first measurement data include camera data from an interior camera, heart rate data from a seat and / or steering wheel sensor and / or skin conductance data from a sensor to detect the skin conductance of the driver. [3] Computer-implemented method (100) according to claim 1 or 2, characterized by that the second measurement data includes steering wheel and / or pedal movement data and / or vehicle dynamics data. [4] Computer-implemented method (100) according to any one of the preceding claims, characterized bythat the situation context agent further uses navigation and / or traffic data, which in particular indicate congestion, weather conditions and / or the complexity of the route, to determine the driving situation. [5] Computer-implemented method (100) according to any one of the preceding claims, characterized by that the fatigue agent is trained to recognize fatigue, microsleep and / or slowed reactions as a health condition. [6] Computer-implemented method (100) according to any one of the preceding claims, characterized by that the stress agent is trained to recognize overload, increased tension and / or situational uncertainty. [7] Computer-implemented method (100) according to any one of the preceding claims, characterized by, that the intervention agent is trained to provide at least one warning signal, one control signal for an interior vehicle function (18), one start signal for at least one assistance system and / or to steer the vehicle (10) to the nearest parking space. [8] Computer program product comprising instructions which, when the program is executed by a computer (12), cause it to perform the steps of the method (100) according to any one of claims 1 to 7. [9] Vehicle (10) comprising at least one vehicle function (18), a first sensor device (14) for acquiring first measurement data indicating a health value of a vehicle driver, and a second sensor device (16) for acquiring second measurement data indicating a driving situation of the vehicle (10) while driving, and comprising at least one computer (12) configured to receive the first measurement data, the second measurement data and to perform the steps of the method (100) according to any one of claims 1 to 7.

Images