US-12617252-B2 - Method for setting an air conditioner

US12617252B2US 12617252 B2US12617252 B2US 12617252B2US-12617252-B2

Abstract

A method for individualized setting of an air conditioner in a vehicle for a user using a control architecture is provided. Data relating to setting the air conditioner is acquired for all users and all vehicles and an average basis model is trained in an AI unit using the data acquired for all users and all vehicles. A comfort model is then created in the AI unit from at least the basis model and individualized settings for the air conditioner for a specific user are predicted based on the comfort model. A control architecture for executing the method is also provided.

Inventors

Marcel Günter
Marco Roth
Kristian Haase
Lars Ludwig

Assignees

MAHLE INTERNATIONAL GMBH

Dates

Publication Date: 20260505
Application Date: 20240625
Priority Date: 20230627

Claims (10)

1 . A method for individualized setting of an air conditioner in a vehicle for a user, wherein, by means of a control architecture: data relating to setting the air conditioner is acquired for all users and/or all vehicles, an average basis model is trained in an AI unit using the data acquired for all users and all vehicles, a comfort model is created in the AI unit by: superimposing a vehicle model of a specific type of vehicle, a user model for a specific user of numerous vehicles, and a user/vehicle model for a specific user of a specific type of vehicle on the average basis model, wherein the vehicle model, user model, and user/vehicle model differ from each other; and controlling an air conditioner for a specific user based on the comfort model and an individualized prediction.
2 . The method according to claim 1 , wherein the average basis model is obtained with a performance map model.
3 . The method according to claim 1 , wherein at least one sub-model is also trained in the AI unit and the comfort model is created by superimposing the at least one sub-model on the comfort model.
4 . The method according to claim 3 , wherein the respective sub-model is a variation of the average basis model.
5 . The method according to claim 3 , wherein the sub-model is trained by means of the control architecture as a vehicle model, wherein the vehicle model is a model based on the data acquired for all users of a specific type of vehicle.
6 . The method according to claim 3 , wherein the sub-model is trained by means of the control architecture as a user model, wherein the user model is a model based on the data acquired for a specific user of numerous vehicles.
7 . The method according to claim 3 , wherein the sub-model is trained by means of the control architecture as a user/vehicle model, wherein the user/vehicle model is a model based on the data acquired for a specific user of a specific type of vehicle.
8 . The method according to claim 5 , wherein the vehicle model and/or a user model, and/or a user/vehicle model differ from one another.
9 . The method according to claim 5 , wherein the comfort model is created by superimposing the vehicle model, and/or a user model, and/or a user/vehicle model on the average basis model.
10 . A control architecture for executing the method according to claim 1 , wherein the control architecture is designed to: acquire data relating to setting the air conditioner for all users and all vehicles, train an average basis model based on the data acquired for all users and all vehicles in an AI unit, create a comfort model from at least the average basis model in the AI unit, predict individualized settings for the air conditioner for a specific user based on the comfort model.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority from German Patent Application No. 10 2023 206 038.8, filed on Jun. 27, 2023, the entirety of which is hereby fully incorporated by reference herein. The invention relates to a method for individualized setting of an air conditioner in a vehicle using a control architecture according to the preamble of claim 1. The invention also relates to the control architecture for executing the method. An air conditioner user can set the temperature by changing numerous parameters based on the user's thermal perceptions. The air conditioner can be set by a climate control system based on an engine performance map, thus simplifying use of the air conditioner. The thermal perceptions of an individual user may be very specific, however. Conventional performance map-based climate controls are configured for “average” people, and cannot be automatically adjusted to individual users. If the user does not feel comfortable with performance map-based climate control, the air conditioner has to be adjusted manually. AI-supported comfort models (AI: artificial intelligence) that can be adjusted to individual preferences have also been proposed. With these, the user's settings for the air conditioner are correlated with context data, e.g. exterior temperature or interior temperature, and used for training the comfort model. The trained comfort model can then predict the optimal settings of the air conditioner for individual users at predefined times from the context data. Unfortunately, the comfort model cannot be used for other users and other vehicles with this solution. The object of the invention is to therefore create an improved, or at least alternative, method with which the above disadvantages are resolved. It is also the object of the invention to create a corresponding control architecture for executing the method. This is achieved according to the invention by the subject matter of the dependent claims. Advantageous embodiments are the subject matter of the dependent claims. The present invention is based on the general idea of creating the comfort model through a superimposing different models on one another, so that it can be used for different vehicles and different users. The method is intended for individualized setting of an air conditioner in a vehicle for a user using a control architecture. Data relating to setting the air conditioner is acquired by the control architecture for all users and/or all vehicles, and an average basis model is trained with this data in an AI unit. A comfort model is then created in the AI unit from at least the basis model, and individualized settings for the air conditioner are then predicted for a specific user based on the comfort model. The comfort model therefore contains at least the average basis model, which takes the data acquired for all users and all vehicles into account. The trained basis model can therefore be used to predict individualized settings for all users in all vehicles, or for specific users in specific vehicles. Accordingly, the comfort model based on the basis model can be used for any vehicle and/or any user. Consequently, the comfort model can reduce the amount of manual input from the user, and it can be used for any vehicle and/or any user with less difficulty. The basis model and comfort model can be based on data, which reduces the development difficulties significantly in comparison with performance map-based models. Alternatively, the basis model can also be obtained using a performance map-model. Advantageously, the data acquired for all users and all vehicles can be sent to a cloud storage, and the average basis model can be trained with this data in the AI unit in the cloud. The comfort model can then also be created from at least the average basis model in the cloud. Consequently, the average basis model and/or a sub-model, described in greater detail below, can be trained in a central location using the data acquired for all users and all vehicles. The trained comfort model can then be sent from the cloud to a specific vehicle that contains the air conditioner that is to be set, and the individualized settings for the air conditioner for a specific user can be sent directly to the vehicle. Alternatively, the individualized settings for the air conditioner in a specific vehicle for a specific user can be determined in the cloud and then sent to the vehicle. The cloud is a network that provides computer resources such as servers, and/or data storage, and/or applications. In the context of the present invention, vehicles are any vehicles that have air conditioners, regardless of their properties. The vehicles can be of different types, each type of which comprises numerous vehicles with identical properties. A specific vehicle may be classified as a specific type. These vehicles can be from a specific manufacturer, for example, or they can be specific models of vehicles from a