DE102022200511A1 - Automatic control of a household appliance - Google Patents

Abstract

According to a method for automatically controlling a household appliance (3) for storing or preparing food depending on the movement of a user (5) in an environment (1) of the household appliance (3), radio signals are exchanged between at least two transceivers (8, 9) for wireless communication via a radio network, the at least two transceivers (8, 9) being positioned in relation to the household appliance (3) in such a way that the radio signals propagate at least partially through the environment (1) of the household appliance (3). At least one intrinsic parameter of the exchanged radio signals, which is influenced by the movement of a user (5) in the area (1) of the household appliance (3), is determined as a function of time and based on the at least one time-dependent intrinsic parameter, a predefined behavior pattern of the user (5) is identified. The household appliance (3) is automatically controlled depending on the behavior pattern identified.

Description

The invention relates to a method for automatically controlling a household appliance for storing or preparing food depending on the movement of a user in the vicinity of the household appliance, a corresponding control system for a household appliance and a household appliance for storing or preparing food with such a control system.

The networking of electrical household appliances, such as stoves, ovens, refrigerators or other cooling devices, in the technological context of the so-called intelligent home (English: "Smart Home") or the so-called Internet of Things, loT (English: "Internet of Things"), can be used for a variety of applications. For this purpose, household appliances are equipped with WLAN modules, for example.

In order to be able to use the resulting possibilities as well as possible, it would be desirable to be able to detect user actions in the spatial environment of the household appliance in order to automate the control of the household appliance more comprehensively. For example, radar sensors can be used for simple user actions, such as the user approaching the household appliance. However, more complex user behavior cannot be reliably identified in this way.

Furthermore, techniques for recognizing persons, their movement and gestures and so on, based on influencing radio signals, the person not having to carry any transmitting or receiving devices, are known per se. For example, it has been proposed to monitor a person's breathing in this way while they are sleeping. For reviews of such techniques, referred to for example as WiFi motion detection technologies or device free wireless sensing (DFWS), see the publications by J. Wang et al., for example: " Device-Free Wireless Sensing: Challenges, Opportunities, and Applications”, IEEE Network 32, 2, 132 (2018), and R. Zhang et al., Device-Free Human Identification Using Behavior Signatures in WiFi Sensing, Sensors 21, 17, 5921 (2021).

It is an object of the present invention to be able to take into account more complex user behavior in the spatial environment of the household appliances in the automatic control of household appliances for storing or preparing food.

This object is solved by the respective subject matter of the independent claims. Advantageous and preferred developments are the subject matter of the dependent claims.

The invention is based on the idea of analyzing the influence of intrinsic parameters of exchanged radio signals as a function of time and inferring user behavior or a predefined behavior pattern of the user from this, on which the control of the household appliance is based.

According to one aspect of the invention, a method for automatically controlling a household appliance depending on the movement of a user in an area surrounding the household appliance is specified, the household appliance being a household appliance for storing or preparing food. According to the method, radio signals are exchanged between at least two transceivers, which can also be referred to as transceivers or transceivers, for wireless communication via a radio network. The at least two transceivers are positioned in relation to the household appliance in such a way that the radio signals propagate at least partially through the surroundings of the household appliance. At least one intrinsic parameter of the exchanged radio signals, which can be influenced or will be influenced by the movement of a user in the vicinity of the household appliance, is determined as a function of time, in particular by means of a computing unit of a control system for the household appliance. Based on the at least one time-dependent intrinsic parameter, ie the at least one intrinsic parameter determined as a function of time, a predefined behavior pattern of the user is identified, in particular by means of the computing unit. The household appliance is controlled automatically depending on the identified behavior pattern, in particular by means of a control unit of the control system or by means of the computing unit.

The household appliance is in particular an electrically operated or electrically operable household appliance. If the household appliance is designed for storing food, it is designed in particular for temperature-controlled or refrigerated storage of food, for example as a cooling device, in particular as a refrigerator and/or freezer, chest freezer or wine refrigerator. If the household appliance is designed for preparing food, it is designed in particular as an oven and/or hob or stove.

Under a transceiver for wireless communication over a radio network can be a device be understood, which is designed to transmit and receive corresponding radio signals. Such transceivers can have, for example, a receiving antenna and a transmitting antenna or a combined receiving and transmitting antenna, as well as a front-end circuit and optionally one or more further processing circuits.

The radio network is, for example, a local radio network, for example a WLAN network (Wireless Local Area Network). The transceivers are then correspondingly WLAN transceivers. The use of a radio network such as WLAN, which is used anyway for communication with the devices for other purposes, is particularly advantageous.

In particular, exchanging the radio signals includes the transmission of a first radio signal by a first transceiver of the at least two transceivers to a second transceiver of the at least two transceivers and the detection of the reception of the radio signal by the second transceiver. The at least one intrinsic parameter can be determined based on the received radio signal. The first transceiver can in particular be provided externally to the household appliance, for example as a WLAN router, whereas the second transceiver is part of the household appliance. However, other constellations are also possible.

If the first transceiver is in the form of a WLAN router, for example, it can send a radio signal, also referred to as a broadcast signal, which contains, for example, LTFs (long training symbols). The second transceiver receives the broadcast signal and determines the at least one intrinsic parameter based thereon, for example from the LTFs.

Alternatively, the second transceiver, after having received the radio signal, can send another radio signal back to the first transceiver or to a further transceiver of the at least two transceivers. After the first transceiver or the further transceiver has received this radio signal, the at least one intrinsic parameter can be determined based thereon. The radio signals can be, for example, dedicated test signals that are exchanged specifically for the purpose of being able to detect the movement of the user based on the intrinsic parameters. In addition, the at least two transceivers can also exchange other data or information. Alternatively, the radio signals can be those for conventional communication or for the exchange of data or information, which are also used according to the invention to determine the at least one intrinsic parameter.

The radio signals are exchanged in particular over a predetermined detection period, so that the time-dependent change in the radio signal and accordingly the time-dependency of the at least one intrinsic parameter can be detected. The recording period can also be a sliding period or a sliding time window.

The radio signals can propagate directly between the at least two transceivers and/or reflections, diffraction effects and so on can occur.

Depending on the application and depending on the configuration of the household appliance, controlling the household appliance can include different actions, in particular actions for control in the narrower sense or for regulation. For example, controlling the household appliance can include changing or adjusting or regulating one or more operating parameters of the household appliance. The operating parameters can contain, for example, a temperature and/or air humidity in a storage area for storing the food, or their control or regulation. Alternatively or additionally, controlling the household appliance can control, regulate, activate or deactivate one or more electronic and/or mechanical components of the household appliance, for example a fan, a motor, an actuator, for example an actuator to support opening a door, suction and so on further, include.

The behavior pattern is in particular one of a large number of predefined behavior patterns. Based on the at least one time-dependent intrinsic parameter, it is therefore recognized which of the predefined behavior patterns is present. This is then used as a basis for the automatic control of the household appliance. For example, various measures or actions that the control system initiates can be provided, depending on which of the behavioral patterns has been identified.

The behavior pattern can be characterized, for example, by a trajectory of the user in the area, the time-dependent distance of the user to the household appliance or to another household appliance or to other places in the area, a movement speed, a movement profile, and so on. For example, the motion profile may include the user's position as a function of time of the trajectory, i.e. a time-dependent change in the movement speed, pausing or stopping the user at certain points in the environment, and so on. Depending on the embodiment of the method, such characteristic properties, such as the movement profile and/or the trajectory, can first be determined and, in a second step, the behavior pattern can be identified based on the characteristic properties. In other embodiments, the behavior pattern can be determined directly from the at least one time-dependent intrinsic parameter, for example using a classification algorithm, which can be based in particular on a model trained using machine learning.

Identifying the behavior pattern can mean, for example, that it is determined that the user is apparently performing a specific activity, such as storing groceries, preparing food, cooking, and/or moving towards a refrigerator with the intention of storing something there or to be taken, and so on.

The at least one intrinsic parameter is calculated or determined, in particular by means of the computing unit, based on the exchanged radio signals, ie in particular based on at least one of the exchanged radio signals after it has been received by the corresponding transceiver of the at least two transceivers.

The at least one intrinsic parameter can include any parameters that can be derived from the exchanged radio signals and that can potentially be influenced by the movement. The at least one intrinsic parameter may also include parameters that may be affected by the user's presence in the environment and/or the user's size. In particular, it is known from known applications of device-free detection technologies, such as DFWS, that channel state information, CSI (English: "Channel State Information"), received signal strength or signal propagation time of the exchanged radio signals can be influenced in a characteristic way by the movement and / or presence of a user . The at least one intrinsic parameter can contain such parameters, for example.

The invention thus uses the influence of the radio signals that are exchanged between two corresponding transceivers by the user in the environment in order to recognize behavioral patterns, for example movement in a room, entering the room and / or the targeted movement of the user in a specific direction or toward the household appliance or other specific device. The limitations of simple sensors, such as radar sensors for distance measurement, cannot do this. Since a large number of sidebands, also referred to as sub-carriers, are usually used when radio signals are exchanged via a radio network, a large number of signals or channels and CSI are available for the at least one intrinsic parameter, which are used to derive the at least an intrinsic parameter and can be evaluated with regard to the behavior of the user.

According to at least one embodiment of the method according to the invention, a movement profile of the user is determined based on the at least one time-dependent intrinsic parameter and the behavior pattern is identified as a function of the movement profile.

The movement profile can contain coordinates, in particular at least two-dimensional coordinates, of a position of the user in the vicinity of the household appliance as a function of time, ie it can describe the movement of the user along a trajectory.

The behavior pattern can be identified by the movement profile, with additional information being able to be taken into account in various embodiments, such as camera images, day of the week, time of day, and so on.

Purely as an example, the user who is storing his purchases after returning from shopping could enter a room, which can represent the area around the household appliance, through a specific door, first stop at a specific point to put down bags, bags or boxes, and then move towards a shelf or drawer and then towards the appliance and so on. On the other hand, if the user is preparing food, he could possibly be in the vicinity of a hob or other work surface for a longer period of time. These different movement patterns are reflected in the at least one time-dependent intrinsic parameter.

In various embodiments, the movement profile includes a distance between the user's position in the vicinity of the household appliance and a predefined reference point as a function of time.

The reference point can, for example, correspond to the position of the household appliance or one of the transceivers. Determining the distance requires less complex analysis steps than determining two-dimensional coordinates. If a correspondingly lower information content can be accepted, such embodiments offer corresponding advantages.

According to at least one embodiment, a classification algorithm, for example a classification algorithm trained using machine learning, is applied to input data, the input data containing the at least one time-dependent intrinsic parameter or being determined as a function of the at least one time-dependent intrinsic parameter in order to identify the behavioral pattern.

In other words, identifying the behavior pattern involves applying the classification algorithm to the input data. The behavior pattern can then be given directly as an output of the classification algorithm or can be uniquely determined based on it.

In particular, the specific abilities of machine-trained classification algorithms for classification without concrete knowledge of the underlying meaning of the signatures in the input data can be exploited in order to directly identify the behavioral pattern based on the time-dependent intrinsic parameters, without having to use potentially inaccurate models to first determine movements from the derive intrinsic parameters and, based on them, conclude the behavioral pattern. Furthermore, such embodiments have the advantage that the classification algorithm can be trained or configured in a user-specific or application-specific manner, so that increased reliability can be achieved when identifying the behavior pattern.

Various architectures can be used as classification algorithms, in particular those that can handle time-dependent input data, such as known cluster algorithms, algorithms based on decision trees or algorithms based on reinforcement learning or deep learning. ) are based, are used, so in particular trained artificial neural networks, such as recurrent neural networks, RNN (English: "recurrent neural network"), for example long-short-term memory networks, LSTM (English: "long short-term memory").

The selection of a suitable algorithm from the large number of known classification algorithms that can be trained using machine learning or can be created conventionally without machine learning depends on the specific implementation of the method and in particular on the type of control and the type of controlled household appliance.

In particular, the classification algorithm is also trained in a known manner. For this purpose, for example, test persons can implement certain behavioral patterns, the at least one intrinsic parameter can be determined as a function of time as described, and in this way, for example, monitored learning can be implemented.

According to at least one embodiment, a geometric description of the environment and/or position information, which includes a position of the household appliance in the environment and/or a position of at least one other electrical device and/or at least one other device and/or at least one workplace in the environment, predetermined. The trained classification algorithm is configured depending on the geometric description and/or the position information, in particular before the at least one intrinsic parameter is determined.

In other words, the classification algorithm can be generically pre-trained and adapted to the specific application situation by configuration, ie by specifying appropriate parameters or hyperparameters of the algorithm.

According to at least one embodiment, user information related to the identified behavior pattern is output to the user and user feedback from the user is collected in response to the output of the user information. At least one parameter of the trained classification algorithm is adjusted depending on user feedback.

In this way, the training of the classification algorithm and/or the configuration of the classification algorithm can be refined or improved or adapted in a user-specific or application-specific manner.

The user information therefore contains in particular the information as to which of the behavior patterns is present according to the result of the classification algorithm based on the at least one intrinsic parameter. This information can be given to the user, for example, acoustically and/or which are output visually, for example via a corresponding output device or display device of the household appliance or a mobile electronic device of the user, for example a cell phone. For this purpose, in particular, the processing unit can generate corresponding signals and/or information and transmit them to the output device and/or display device.

The user feedback can then, for example, contain information as to whether the identified behavioral pattern is correct or correct or not. Depending on the embodiment, the user feedback can also contain further information, for example regarding which activity was actually performed if the behavior pattern identified is not the correct behavior pattern.

The user can enter the user feedback via an input device of the household appliance or the mobile electronic terminal device, so that the user feedback can be transmitted accordingly to the processing unit.

According to at least one embodiment, the at least two transceivers are part of the radio network, in particular a WLAN network or a WPAN network (Wireless Private Area Network) or a low-energy wide area network, also known as an LPWAN network (Low Power Wide Area Network”). The radio network can also be a LoRaWAN or a radio network operating in an 868 MHz frequency band, according to LoRaWAN or another protocol.

According to at least one embodiment, a first radio signal of the radio signals exchanged is transmitted from a first transceiver of the at least two transceivers to a second transceiver of the at least two transceivers. In response to the second transceiver receiving the first radio signal, a second radio signal of the exchanged radio signals is transmitted from the second transceiver to the first transceiver. The at least one intrinsic parameter is determined as a function of the second radio signal transmitted to the first transceiver and received by the first transceiver.

In this case, the first and second transceivers are positioned in particular in relation to one another and in relation to the area surrounding the household appliance in such a way that the first radio signal and the second radio signal propagate at least partially through the area surrounding the household appliance.

The computing unit can be coupled or connected to the first transceiver, for example, in order to determine the at least one intrinsic parameter based on the received second radio signal. This has the advantage that the second transceiver does not necessarily have to be adapted or provided specifically for the method. Rather, the second transceiver can be any transceiver that is already present in the vicinity or outside the vicinity of the household appliance, for example a WLAN router. The first transceiver can be part of the household appliance, for example, ie it can be installed in or on the household appliance. The second transceiver is arranged in particular externally to the household appliance. For example, the second transceiver is designed as a network router, in particular as a WLAN router. In other embodiments, however, the second transceiver is part of the household appliance and the first transceiver is arranged externally to the household appliance.

According to at least one embodiment, a multiplicity of first radio signals are transmitted from the first transceiver to a multiplicity of second transceivers of the at least two transceivers and, after receiving the respective first radio signal, each of the second transceivers transmits a corresponding second radio signal to the first transceiver and/or at least one other transceivers of the at least two transceivers. The at least one intrinsic parameter is then determined as a function of all of the second radio signals received.

This increases the coverage of the area by the radio signals, so that the movement of the user in the area can be determined with greater reliability or more precisely, or is reflected more clearly in the at least one intrinsic parameter.

In other embodiments, more complex transmission protocols are used between the first and second plurality of transceivers to create even greater coverage.

According to at least one embodiment, the at least one intrinsic parameter contains at least one item of channel state information, CSI (Channel State Information), of the radio signals exchanged, or the at least one intrinsic parameter is determined as a function of the at least one CSI.

CSI are known channel properties of the radio signals or the communication link implemented by the radio signals. For example, the CSI may include information that describes how a radio signal propagates from the transmitting transceiver to the receiving transceiver and reflects the effect by effects such as scattering, signal power drop, fading and so on.

In various embodiments, the at least one intrinsic parameter, for example the CSI, contains at least one amplitude and/or at least one phase information item of the exchanged radio signals.

According to at least one embodiment, the at least one intrinsic parameter includes at least one received signal strength, RSS (received signal strength), of the radio signals exchanged and/or at least one signal propagation time, TOF (time of flight) of the radio signals exchanged or will determined as a function of the at least one received signal strength and/or the at least one signal propagation time.

Known methods for device-free motion and gesture recognition, in particular DFWS methods, use CSI, RSS, TOF, amplitude and phase information for reliable.

According to at least one embodiment, an actuator of the household appliance is controlled depending on the identified behavior pattern in order to move a mechanical component of the household appliance.

The component of the household appliance can be moved, for example, in order to assist the user in opening a door of the household appliance by motor, for example if the household appliance is a cooling appliance, such as a refrigerator or a freezer, in which there may be a negative pressure in the interior prevails, which is caused, for example, by the cooling of warm air introduced. Alternatively or additionally, the negative pressure can be generated intentionally, for example to prevent the door from being opened unintentionally. Based on the identified behavior pattern, the computing unit can determine, for example, that the user is about to open the door and support this accordingly by moving the component.

The component that is moved can also be one or more slats or covers or flaps or the like that can fluidly separate or connect different spatial areas within a storage area for storing food of the household appliance. This is also particularly useful if it is a refrigeration appliance as a household appliance. Such flaps, slats or other covers can act, for example, between different climatic zones within a refrigerator, for example between a regular storage area for food and a so-called zero-degree compartment or the like. It can be advantageous to control or regulate the air humidity inside a defined climatic zone, in particular depending on what type of food is in it. If, according to the behavior pattern, the user is about to open the door, the computing unit can close the slats, flaps and so on as a precaution in order to prevent an increased entry of moisture into the climatic zone or the like.

According to at least one embodiment, a fan in a storage area for storing the food of the household appliance is controlled depending on the identified behavior pattern.

The fan can be used, for example, to generate the negative pressure mentioned above inside the household appliance. If the user is about to open the door, the negative pressure can be reduced by controlling the fan, for example, in order to make opening the door easier. Alternatively, the fan can be used to generate excess pressure to facilitate opening the door.

According to at least one embodiment, a device of the household appliance for regulating humidity and/or a device of the household appliance for regulating temperature in the storage area for storing the food is controlled or regulated depending on the identified behavior pattern.

For example, target specifications for humidity regulation and/or temperature regulation inside the household appliance, ie in particular inside the refrigerator, or in certain parts of the refrigerator, for example the climate zone, can be changed depending on the behavior pattern. Alternatively or additionally, for example, a specification for the rate of change of the air humidity and/or temperature can also be adjusted as a function of the identified behavioral pattern. For example, situations can be taken into account in which the user is expected to extensively fill, empty or reorganize the storage area.

If, according to the behavior pattern identified, it is to be expected, for example, that only minor changes will be made to the loading of the refrigeration appliance, then the humidity and/or temperature regulation cannot be changed, or not significantly so. If, on the other hand, it is to be expected that the loading situation of the cooling device will change significantly, the humidity and/or temperature regulation can be adjusted more strongly be initiated or released. Corresponding control information for humidity and/or temperature regulation can be obtained, for example, via an external signal, for example based on camera-based object recognition.

According to at least one embodiment, based on the at least one time-dependent intrinsic parameter, identity information relating to an identity of the user or a user group to which the user belongs is determined and the household appliance is automatically controlled depending on the identified behavior pattern and depending on the identity information.

Based on the intrinsic parameter, for example, it can be determined with a certain probability whether the user is a child or an adult, for example because people of different body sizes affect the intrinsic parameters differently. Specific movement patterns of individual users or user groups can also affect the intrinsic parameters. In this way, the identity of the user can at least be narrowed down, particularly if a certain selection of possible users is specified.

The domestic appliance can therefore be controlled individually or in a user-group-specific manner, so that specific preferences or restrictions relating to specific users or user groups can be implemented.

For example, access to or operation of the household appliance can be deactivated or blocked if the user is a specific person or a user from a specific user group, for example a child. For example, a wine cooler can automatically lock or a stove or oven can be deactivated when a child approaches, but not when an adult approaches.

According to at least one embodiment, at least one operating element is blocked, deactivated, unlocked or activated depending on the identified behavior pattern and/or depending on the identity information.

In other words, automatically controlling the home appliance includes locking, disabling, unlocking, or activating the home appliance.

According to a further aspect of the invention, a control system for a household appliance for storing or preparing food is specified. The control system has a first transceiver for wireless communication via a radio network, the first transceiver being set up to exchange radio signals with at least one second transceiver for wireless communication via the radio network. The first transceiver is positioned or can be positioned with respect to the household appliance in such a way that the radio signals propagate at least partially through the surroundings of the household appliance. The control system has a computing unit which is connected or coupled to the first or the second transceiver, for example, and which is set up to assign at least one intrinsic parameter of the exchanged radio signals, which is influenced by the movement of a user in the vicinity of the household appliance, as a function of time determine. The computing unit is set up to identify a predefined behavior pattern of the user based on the at least one time-dependent intrinsic parameter. The control system has at least one control unit that is set up to automatically control the household appliance depending on the identified behavior pattern.

The arithmetic unit and the at least one control unit can be designed in combination with one another or separately from one another.

The first transceiver is part of the control system, whereas the second transceiver can be part of the control system or can be external to the control system, depending on the embodiment of the control system. Furthermore, the control system can include the household appliance and/or the household appliance can include the first transceiver.

According to at least one embodiment of the control system, the control system contains the second transceiver and the second transceiver is designed as a network router.

Further embodiments of the control system according to the invention follow directly from the various embodiments of the method according to the invention and vice versa. In particular, a control system can be set up to carry out a method according to the invention or it carries out such a method.

According to a further aspect of the invention, a household appliance for storing or preparing food is specified, which has a control system according to the invention.

In particular, the second transceiver is not part of the household appliance and the accordingly not part of the control system.

In various embodiments of the household appliance, the first transceiver is installed in or on the household appliance.

According to at least one embodiment, the household appliance is designed as a household appliance for storing food and designed as a refrigerator. For example, the cooling device can be a refrigerator or a wine refrigerator or a freezer.

In accordance with at least one embodiment, the household appliance is designed for preparing food and contains in particular a hob and/or an oven.

Does the present disclosure refer to the fact that a component of the control system according to the invention or the household appliance according to the invention, in particular the computing unit or the control unit, is set up, embodied, designed or the like to perform or realize a specific function, to have a specific effect achieve or serve a specific purpose, this can be understood in such a way that the component, beyond the basic or theoretical usability or suitability of the component for this function, effect or this purpose, through appropriate adaptation, programming, physical design and so on further specific and actually capable of performing or realizing the function, achieving the effect or serving the purpose.

A computing unit can be understood in particular as a data processing device that contains a processing circuit. The arithmetic unit can therefore in particular process data for carrying out arithmetic operations. This may also include operations to perform indexed accesses to a data structure, for example a look-up table (LUT). In various embodiments, the control unit of the control system according to the invention can also be a computing unit or part of a computing unit. The at least one control unit of the control system according to the invention can optionally also be designed as at least one computing unit.

In particular, the processing unit can contain one or more computers, one or more microcontrollers and/or one or more integrated circuits, for example one or more application-specific integrated circuits, ASICs (English: “application-specific integrated circuit”), one or more field-programmable gate Arrays, FPGA, and/or one or more single-chip systems, SoC (English: "system on a chip"). The computing unit can also have one or more processors, for example one or more microprocessors, one or more central processing units, CPU, one or more graphics processor units, GPU and/or contain one or more signal processors, in particular one or more digital signal processors, DSP. The computing unit can also contain a physical or a virtual network of computers or other of the units mentioned.

In various exemplary embodiments, the computing unit includes one or more hardware and/or software interfaces and/or one or more memory units.

A memory device can be configured as volatile data storage, such as dynamic random access memory (DRAM), or static random access memory (SRAM), or non-volatile Data memory, for example as a read-only memory, ROM, as a programmable read-only memory, PROM, as an erasable read-only memory, EPROM (erasable read-only memory) ), as electrically erasable read-only memory, EEPROM (English: "electrically erasable read-only memory"), as flash memory or flash EEPROM, as ferroelectric memory with random access, FRAM (English: "ferroelectric random access memory"), as magnetoresistive memory with random access, MRAM (English: "magnetoresistive random access memory") or as phase change memory with random access, PCRAM (English: "phase-change random access memory"), be configured.

Further features of the invention result from the claims, the figures and the description of the figures. The features and combinations of features mentioned above in the description and the features and combinations of features mentioned below in the description of the figures and/or shown in the figures can be included in the invention not only in the combination specified in each case, but also in other combinations. In particular, the invention can also include designs and combinations of features that do not have all the features of an originally formulated claim. In addition, versions and combinations of features of the The invention encompasses that go beyond or deviate from the combinations of features set out in the back references of the claims.

• Fig. eine schematische Darstellung einer beispielhaften Ausführungsform eines erfindungsgemäßen Steuerungssystems beziehungsweise eines erfindungsgemäßen Haushaltsgeräts.

The invention is explained in more detail below using specific exemplary embodiments and associated schematic drawings. It shows:

• Fig. shows a schematic representation of an exemplary embodiment of a control system according to the invention or a household appliance according to the invention.

FIG. 1 shows a schematic plan view of a household appliance 3 according to the invention in a room 1, in particular a living room or a kitchen. The household appliance 3 is designed for storing or preparing food, without restricting the generality, for example as a refrigerator. In the room 1 more jobs and devices 4 may be present. Furthermore, a door 2 for entering the room 1 is shown as an example.

The household appliance 3 has a first WLAN transceiver 8, which can exchange radio signals with a second WLAN transceiver 9, which is arranged outside the household appliance 3, via a WLAN network. The second transceiver 9 is in 1 also located in room 1, but it can generally be located in another room. For example, the second transceiver 9 can be a WLAN router.

In 1 a user 5 is also shown schematically, who can move along different trajectories 6, 7 within the room 1.

While the user 5 is in the room 1, the transceivers 8, 9 exchange radio signals, for example the first transceiver 8 transmits a first radio signal to the second transceiver 9 and the second transceiver 9 in response transmits a second radio signal to the first transceiver 8 sends. Alternatively, the second transceiver 9 sends the second radio signal to the first transceiver 8 without the latter first sending the first radio signal to the second transceiver 9 , that is to say without requesting the second radio signal from the second transceiver 9 .

The household appliance 3 also has a computing unit 10 which is coupled to the first transceiver 8 . The processing unit 10 can determine at least one intrinsic parameter of the second radio signal as a function of time based on the received second radio signal and can identify a predefined behavior pattern of the user 5 based on the time-dependent intrinsic parameter. The household appliance 3, for example the computing unit 10, also has at least one control unit which is set up to automatically control the household appliance 3 depending on the identified behavior pattern.

Optionally, a camera 11 can also be provided, which can capture at least part of the room 1 and can transmit corresponding image data, for example wirelessly via the first transceiver 8, to the computing unit 10. In various embodiments, the computing unit 10 can also take into account the image data in addition to the at least one intrinsic parameter in order to identify the behavior pattern.

Depending on which trajectory 6, 7 and/or at which speed or in which way the user 5 moves along the trajectory 6, 7, the exchange of radio signals between the transceivers 8, 9 is influenced in different ways. This manifests itself in differences in the time-dependent intrinsic parameter. Depending on the analysis of the at least one time-dependent intrinsic parameter, the computing unit 10 can reliably recognize complex user actions, such as stowing away purchases, the preparation of food and meals, as one of several predefined behavior patterns and thereby initiate or control targeted actions of the household appliance 3.

For example, parameters such as a temperature and/or humidity level inside the refrigerator or part of the interior of the refrigerator, for example a zero-degree compartment, could be adjusted specifically according to the identified behavioral pattern. For this purpose, the arithmetic unit 10 can, via the at least one control unit (not shown), adjust the output of a cooling unit of the refrigerator accordingly and/or control a device for regulating humidity, for example using motorized flaps.

Techniques known per se, such as DWFS, can be used in order to be able to analyze a large number of pieces of information, for example different CSI, in a time-dependent manner and to obtain a clear movement profile of the user 5 or the like from this. In particular, correlations of individual OFDM sub-carriers or their CSI can be used or other features of a CSI.

The at least one time-dependent intrinsic parameter can be analyzed, for example, by means of the computing unit 10, so that the time-dependent coordinates of the trajectories 6, 7 can be determined explicitly or a distance of the user 5 from the first transceiver 8 or the like can be determined as a function of time. The behavior pattern can then be determined based on the trajectories 6, 7 or based on the distance.

Alternatively, the at least one time-dependent intrinsic parameter is used as part of input data for a machine-learning-trained classification algorithm that can directly assign the timing and characteristic patterns of the at least one time-dependent intrinsic parameter to a behavioral pattern.

During the training of the classification algorithm, complex user interactions and behaviors can be learned by combining them with temporal usage patterns and, if necessary, other signals, such as the image data. For example, the recognition of the behavioral pattern "putting away purchases" can also be supported by the fact that it occurs frequently on a certain day of the week and/or at a certain time. This accumulation can be derived, for example, from the fact that the user 5 moves according to a specific movement pattern, a door of the refrigerator remains open for a relatively long time and/or the load status of the refrigerator increases significantly.

In some embodiments, the processing unit 10 can control the household appliance 3 depending on the identified behavior pattern, so that a noisy process, for example defrosting the refrigerator or freezer, is carried out in a targeted manner when the user 5 is not in the room 1 .

The classification algorithm for individual users 5 can advantageously be adapted automatically through further learning, since the CSI signals can depend on the installation conditions and the room situation. For example, installation conditions and room situation can influence which correlation patterns of the channels are set. The classification algorithm can be adjusted accordingly via customer interaction or user feedback, in particular as to whether the behavior pattern was identified correctly or incorrectly.

Another application, if the household appliance 3 is a refrigerator with a specific climate zone, such as a zero-degree compartment, the evaluation of the at least one intrinsic time-dependent parameter can be combined with the image data in order to identify the behavioral pattern. For example, the image data can be used to determine how much the load in the refrigerator changes. The camera 11 can be arranged in the immediate vicinity or inside the refrigerator. If, for example, it was recognized as a behavior pattern that the user 5 is about to put his purchases away and it was determined on the basis of the image data that the load in the zero-degree compartment changes significantly, the humidity and/or temperature settings can be automatically adjusted accordingly .

In further embodiments, one or more further transceivers (not shown) are distributed in room 1 or outside room 1 and radio signals are exchanged between the transceivers in order to increase the coverage of room 1 and thus the reliability or accuracy in identifying the behavior pattern .

Reference List

1

Space

2

door

3

household appliance

4

workstations and equipment

5

user

6, 7

trajectories

8, 9

transceivers

10

unit of account

11

camera

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Non-patent Literature Cited

• R. Zhang et al.: "Device-Free Human Identification Using Behavior Signatures in WiFi Sensing", Sensors 21, 17, 5921 [0004]

Claims (15)

Method for automatically controlling a household appliance (3) for storing or preparing food depending on the movement of a user (5) in an environment (1) of the household appliance (3), characterized in that - between at least two transceivers (8, 9) radio signals are exchanged for wireless communication via a radio network, the at least two transceivers (8, 9) being positioned in relation to the household appliance (3) in such a way that the radio signals propagate at least partially through the environment (1) of the household appliance (3); - At least one intrinsic parameter of the exchanged radio signals, which can be influenced by the movement of the user (5) in the area (1) of the household appliance (3), is determined as a function of time; - A predefined behavior pattern of the user (5) is identified based on the at least one time-dependent intrinsic parameter; and - the household appliance (3) is automatically controlled depending on the identified behavior pattern. procedure after claim 1 , characterized in that based on the at least one time-dependent intrinsic parameter, a movement profile of the user (5) is determined and the behavior pattern is identified depending on the movement profile. procedure after claim 2 , characterized in that - the movement profile contains coordinates of a position of the user (5) in the area (1) of the household appliance (3) as a function of time; and/or - the movement profile contains a distance between the position of the user (5) in the vicinity (1) of the household appliance (3) and a predetermined reference point as a function of time. procedure after claim 1 , characterized in that a classification algorithm, in particular a classification algorithm trained by means of machine learning, is applied to input data which contain the at least one time-dependent intrinsic parameter or are determined as a function of the at least one time-dependent intrinsic parameter in order to identify the behavioral pattern. procedure after claim 4 , characterized in that - a geometric description of the environment (1) and/or position information which shows a position of the domestic appliance (3) in the environment (1) and/or a position of at least one further electrical appliance and/or at least one further appliance and /or contain at least one workplace in the area (1), are specified; - the classification algorithm is configured depending on the geometric description and/or the position information. Procedure according to one of Claims 4 or 5 , characterized in that - user information relating to the identified behavior pattern is output to the user (5); - detecting user feedback from the user (5) in response to the issuance of the user information; and - at least one parameter of the classification algorithm is adjusted depending on the user feedback. Method according to one of the preceding claims, characterized in that the at least two transceivers (8, 9) are part of a WLAN network or a WPAN network or an LPWAN network. Method according to one of the preceding claims, characterized in that a first radio signal is transmitted from a first transceiver (8) of the at least two transceivers (8, 9) to a second transceiver (9) of the at least two transceivers (8, 9); and - the at least one intrinsic parameter is determined as a function of the first radio signal transmitted to the second transceiver (9); or - in response to receipt of the first radio signal by the second transceiver (9), a second radio signal is transmitted from the second transceiver (9) to the first transceiver (8) and the at least one intrinsic parameter depends on the signal sent to the first transceiver ( 8) transmitted second radio signal is determined. Method according to one of the preceding claims, characterized in that the at least one intrinsic parameter - contains at least one item of channel status information of the exchanged radio signals or is determined as a function of the at least one item of channel status information; and/or - contains at least one received signal strength of the exchanged radio signals or is determined as a function of the at least one received signal strength; and/or - information that depends on at least one signal propagation time of the exchanged radio signals, contains or is determined as a function of the at least one signal propagation time. Method according to one of the preceding claims, characterized in that - depending on the identified behavior pattern, an actuator of the household appliance (3) is controlled in order to move a mechanical component of the household appliance (3); and/or - depending on the identified behavioral pattern, a fan in a storage area for storing the food of the household appliance (3) is controlled; and/or - depending on the identified behavior pattern, a device of the household appliance (3) for humidity regulation and/or a device of the household appliance (3) for temperature regulation in the storage area for storing the food is controlled or regulated. Method according to one of the preceding claims, characterized in that based on the at least one time-dependent intrinsic parameter, identity information relating to an identity of the user (5) or a user group to which the user (5) belongs is determined and the household appliance (3) depending on the identity information is automatically controlled. Method according to one of the preceding claims, characterized in that at least one operating element of the household appliance (3) is blocked, deactivated, unlocked or activated depending on the identified behavior pattern. Control system for a household appliance (3) for storing or preparing food, having a first transceiver (8) for wireless communication via a radio network, which is set up to exchange radio signals with at least one second transceiver (9) for wireless communication via the radio network, characterized in that - the first transceiver (8) is positioned or can be positioned with respect to the household appliance (3) in such a way that the radio signals propagate at least partially through the environment (1) of the household appliance (3); - the control system has a computing unit (10) which is set up to determine, as a function of time, at least one intrinsic parameter of the exchanged radio signals, which can be influenced by the movement of a user (5) in an environment (1) of the household appliance (3), and identify a predefined behavioral pattern of the user (5) based on the at least one time-dependent intrinsic parameter; and - the control system has at least one control unit which is set up to automatically control the household appliance (3) as a function of the identified behavior pattern. Household appliance (3) for storing or preparing food, which has a control system Claim 13 having. Household appliance (3) after Claim 14 , characterized in that - the household appliance (3) is designed as a refrigerator; - The household appliance (3) is designed as an extractor hood; or - the household appliance (3) contains a hob and/or an oven.

Images