METHOD AND SYSTEM FOR CONTROLLING A COOKTOP

Description

The present invention relates to a method and a system for controlling a cooktop.

These days, in modern kitchens, it is known to control devices by means of voice recognition. In this context, the requisite voice commands must be very technical and detailed in order to generate a corresponding control signal from the voice commands.

One disadvantage is therefore that this type of voice control of the devices by the user is very difficult.

The object of the present invention is therefore to create a solution which permits a simple control of a cooktop.

The idea underlying the invention is that this object can be achieved by a context being determined, to which voice instructions used intuitively by the user generally relate.

According to a first aspect, the invention relates to a method for controlling a cooktop. The method is characterized in that the method comprises the following steps:

• • Detecting an object
  • Determining information relating to the object,
  • Detecting a voice instruction,
  • Processing the detected voice instruction with at least one part of the determined information and
  • Controlling the cooktop on the basis of the processing result.

The setting of a heating level or heating duration of one or more cooking zones of the cooktop is referred to in particular as controlling a cooktop. The control of the cooktop can also be referred to as activation.

The inventive method comprises the step of detecting an object. The object is preferably detected by a visual detection unit, such as a camera, for instance. In this step, the entire cooktop is preferably monitored by the detection unit and a number of objects are thus also possibly detected. With the inventive method, an item of cookware is preferably used as an object.

One further step of the inventive method is determining information relating to the object. The information relating to the object represents in particular information by means of which the object can be brought into context. To this end, information relating to the object itself and preferably also environmental information relating to the object is preferably determined.

The receiving of an acoustic voice instruction on an acoustic receiving unit is in particular referred to as detecting a voice instruction. The detection of a voice instruction also preferably comprises a conversion of the acoustic voice instruction into instruction data. By way of example, instruction data in the form of words can be produced from an acoustic voice instruction, which can be stored digitally.

In a further step of the inventive method, the detected voice instruction is processed with at least one part of the determined information. The processing can comprise a comparison and/or an assignment. In particular, the processing can comprise an alignment with stored information.

The cooktop is controlled on the basis of the processing result. For instance, the heating level and/or heating duration of a cooking zone is adjusted on the basis of the processing result.

As information relating to the object is also taken into account during control of the cooktop in addition to the voice instruction of a user, in accordance with the invention, a context-related control can be created. In particular, the voice control can as a result take place much more easily, naturally and intuitively.

According to a preferred embodiment, the step of detecting an object comprises the implementation of an object recognition. In particular, at least one object can be identified from a detected image by means of object recognition. The object recognition can be executed with the aid of an algorithm, which for instance compares previously marked and stored images with the current image detected by a detection unit, for instance a camera. During object recognition, the object can be identified as an item of cookware and recognized as such, for instance.

According to one embodiment, the determination of information relating to the object comprises detection of at least one object attribute. The object, the object attribute of which is detected, is the object detected and preferably identified in the detection step. A property or a state of the object is referred to in particular as an object attribute.

As at least one object attribute is taken into consideration during control of the cooktop in addition to the voice instruction, a voice instruction of the user can be assigned more easily to a specific object.

Object attributes, which can be detected with an object, which represents an item of cookware, are for instance the position, the content and/or the type of cookware and/or the state of the content. A pot or pan can be detected as a type of cookware, for instance. The position of the object on the cooktop is in particular referred to as the position of the item of cookware. In particular, the position of the item of cookware can be a position of the item of cookware on a cooking zone of the cooktop. The contents of the item of cookware represent the food to be cooked which is to be prepared in the item of cookware. In addition or alternatively, the state of the contents of the item of cookware can also be detected as an object attribute. In this regard, the boiling of water in a pan can be identified, for instance.

The object attributes can preferably be detected by means of an algorithm in which for instance a currently detected image are compared with stored object attributes.

By detecting object attributes, the voice instruction of the user can be assigned more precisely to an object and in doing so can simultaneously reduce the demand on the technical specifications in the voice instruction. Moreover, the object attributes, in particular the position of the item of cookware, can be used at least partially also as information when the cooktop is controlled.

According to a further aspect, the step of determining information relating to the object comprises the detection of at least one operating state of the cooktop. In particular, in this step, the heating degree or the heating level of the cooking zone, upon which the detected object is located, can be detected. This can take place for instance by way of sensors or by way of wireless communication between the cooktop and a detection unit.

According to a preferred embodiment, the step of determining the information comprises combining at least two object attributes. The combination of object attributes can represent the context of the object. For instance, the context “pot with water” can be generated from the information “pot” and “water”.

According to one embodiment, the step of processing the voice instruction with the determined information can take place the identification of an object by means of comparing a stored assignment of the voice instruction to information relating to the object.

For instance, the context “water in pot” can be assigned both to a voice instruction “egg water” and also “noodle water”. If an object is currently detected, the object attributes of which comprise “water in pot” and the voice instruction “boil noodle water” is given by the user, the correct object, in particular the pot with water, can be easily identified by checking the stored assignment and the corresponding instruction can be carried out with respect to this pot.

According to one embodiment, the control of the cooktop represents the activation of the cooking zone, upon which the identified object is located. In the previously cited example, the cooking zone, upon which the pot with water is located, can be controlled in order to bring the water to boil.

According to one embodiment, the voice instruction represents a target specification of a cooking process. For instance, the target information “please make a hard-boiled egg” may sound. If a pot with water with an introduced egg has been identified on a cooking zone by detecting the object and the object attributes, the cooking zone, upon which the pan has been identified, can be activated with this target information. Moreover, a heating duration of the cooking zone can be set by the target information in addition to a heating level.

According to one embodiment, the method comprises the step of identifying a user. The voice of the user is preferably used to identify the user. The control of the cooktop is preferably carried out as a function of the user. With the instruction “boil an egg”, the cooking state preferred by the user and the cooking duration associated therewith can automatically be set, for instance. For this purpose, the user can store a user profile. The data relating to the identified user can be read out from this user profile upon identification of the user.

According to one embodiment, the method can comprise the step of outputting information to the user. In particular, the information can be a question or confirmation inquiry. In particular, a question or confirmation inquiry can be output when a number of objects are detected on the cooktop, the object attributes of which do not permit a unique assignment to the received voice instruction. When two pots with water are placed on the cooktop for instance and if the voice instruction “boil noodle water” is detected, this voice instruction cannot be assigned uniquely to one of the pots. In this case, a question can be posed to the user in which for example the detected positions of the two pots can be specified for selection. The user can respond by means of a voice input.

According to a further aspect, the invention relates to a system for controlling a hotplate, which is characterized in that this is designed to carry out the inventive method.

Advantages and features which were and are described with respect to the inventive method apply, if applicable, also to the inventive system and vice versa.

According to one embodiment, the system comprises a detection unit for detecting objects, a determination unit for determining information relating to a, in particular the, detected object, a receiving unit for detecting a voice instruction and a processing unit for processing the detected voice instruction with at least one part of the information relating to the object and a control unit for controlling the cooktop on the basis of the processing results.

The units of the system can be integrated at least partially with one another in a shared unit. The units of the system can be designed at least partially as software.

The detection unit for detecting objects preferably represents a visual detection unit, in particular a camera. The detection unit can be arranged for instance on an extractor hood or as a self-sufficient device at a corresponding height above the cooktop. The detection unit can also be designed to carry out an object recognition. Alternatively, the detection unit can be connected to a processing unit, in which the object recognition can be carried out.

The determination unit for determining information relating to an object can be part of the detection unit or a processing unit connected thereto for the object recognition. In addition or alternatively, the determination unit can be connected to a control unit of the cooktop.

The receiving unit for detecting a voice instruction can be provided on the cooktop itself, on the detection unit or on an extractor hood, for instance.

The processing unit for processing the detected voice instruction with at least one part of the information relating to the object can be integrated in the detection unit. Alternatively, the processing unit can be provided in a separate device which is connected or networked with the detection unit, for instance a mobile terminal or a server.

The control unit for controlling the cooktop is used in particular to generate control information and/or a control signal. The control unit can be integrated in the control of the cooktop. Alternatively, the control unit can also be integrated in the detection unit or a processing unit.

According to one embodiment, the system comprises a storage unit for storing detection results, correlations between voice instruction and object information and/or user profiles.

If the units of the system are not provided in a shared unit, these are however designed for communication, in particular wireless communication, with one another.

The invention is described in more detail below with reference to the appended figures, in which:

FIG. 1 shows a schematic block diagram of an embodiment of an inventive system; and

FIG. 2 shows a flow chart of an embodiment of the inventive method.

As FIG. 1 shows, the system 1 comprises a detection unit 10, a receiving unit 11, a processing unit 12, a determination unit 13 and a control unit 14. In the embodiment in FIG. 1, the system 1 moreover comprises a storage unit 15.

In the embodiment shown in FIG. 1, the system 1 is provided on an extractor hood 2 or integrated herein. In the embodiment shown, the detection unit 10 is arranged on the lower side of the extractor hood 2. The detection unit 10 preferably comprises a camera.

A cooktop 3 is arranged below the extractor hood 2. The detection region of the detection unit is aligned in the direction of the cooktop 3 and indicated schematically by the dashed lines in FIG. 1. The size of the surface of the detection region in the plane of the cooktop 3 corresponds here to the surface of the cooktop 3. In the embodiment shown, the detection region of the detection unit 10 therefore covers the cooktop 3.

If cooking takes place on the cooktop 3, the detection unit 10 preferably detects the entire cooking process. During this a video recording is taken throughout the entire cooking process. In this regard, objects in the detection region of the detection unit 10 are in particular detected.

If a pot is placed on the cooktop 3 for instance, this is located in the detection region of the detection unit 10.

The pot can be identified as an object by an object recognition in the detection unit 10 or a recognition unit (not shown). Moreover, object attributes relating to the pot can be determined in the determination unit 13. To this end, the image detected by the detection unit 10 can be transferred to the determination unit 13. For instance, the contents of the pot can be determined in the determination unit 13. The detection unit 10 and the determination unit 13 can be connected to the storage unit 15.

A voice instruction of the user of the cooktop 3 can be received by way of the receiving unit 11. This voice instruction can be processed in the receiving unit 11 or in the processing unit 12, in particular converted. The voice instruction can be processed with the determined object attributes in the processing unit 12. On account of the processing result, a control instruction is generated in the control unit 14 and transmitted to the cooktop 3.

FIG. 2 shows a schematic representation in a flow chart of an embodiment of the inventive method.

The camera preferably provided in the detection unit 10 is used to constantly analyze the events on the cooktop 3 and identify items of cookware (pans, pots etc.) and possibly their positions. Moreover, it is also identified where possible what is currently being prepared in the pot or pan. These attributes are preferably stored continuously. If the cooktop 3 has a data exchange/interface, the system 1 can also query the settings of the cooktop 3. The stored attributes and the settings of the cooktop 3 represent a context for a voice input, in particular voice instruction. The system not only knows where a pot is located, but also the context relating to this pot. On account of this context, simplified speech commands can be used by the user.

From the combination, “position=top right, “contents=pasta+water”, “type=pot”, “noodle water” is thus produced as a simplified context-based voice label of the pot. For the oral communication, the user can now use the term “noodle water”, instead of the term “pot top right”. On account of the voice instruction “boil noodle water”, the cooking zone, upon which the pan is standing (in the example: front right) can be activated so that its heating level is increased.

The present invention is not restricted to the embodiments shown.

In a further scenario, the user provides the instruction for instance “bring the water in the pan to boil”. On account of the context, the system knows where the pan is standing and can control the cooktop accordingly. Moreover, the pan is now also monitored and there is optionally the possibility of reducing the heat once the water is boiling.

If the system has identified a number of pots, for instance, it can request a more precise description from the user by means of questions.

Moreover, the context-sensitive voice input can also be extended arbitrarily according to the invention. The user could also for instance place an egg in the water and say: “Please make a hard-boiled egg”. The system can therefore determine the position of the pot, setting for the cooktop and also cooking time automatically.

Moreover, it is possible to assign user profiles to the voice by means of recognition, for instance with preferred recipes or cooking levels. The following command can thus also be implemented: “Make my usual breakfast egg”.

A content-related or target-oriented control is thereby possible by means of the monitor information which, from the detection unit, which can represent a networked camera system. By means of the camera system, an object on the cooktop can be analyzed and categorized and according to the situation a treatment space can thus be provided by means of the inventive system.

One possibility is therefore produced with the invention which renders the voice input for controlling a cooktop or possibly also an extractor hood or other devices as simple and intuitive as possible. To this end, a context-related control is enabled by the invention.

The present invention has a series of advantages. In particular, complex voice instructions such as “set cooktop plate number 3 to heating level 4”, which are required in the prior art, can be avoided. With the invention, in which the voice instruction is processed by taking a context into account, a simple voice instruction such as “start cooking pasta” is sufficient.

With the present invention, the voice control for the kitchen can therefore take place much more easily, naturally and intuitively than with previous solutions, by the context being known as a result of the image processing and being included in the processing of the speech. The control of the cooktop can also be defined here by the wording of a target state. Without the context, a voice control can only process individual method steps.

Previous voice systems such as for instance in the case of an oven or a coffee machine are sufficient for singular command controllers and sequences. With multiple control elements and events (operation of several cooktops and items of cookware), a unique context-related command assignment and processing is the prerequisite to ensuring a rapid and specific control. This is possible with the present invention.

LIST OF REFERENCE CHARACTERS

• • 1 system
  • 10 detection unit
  • 11 receiving unit
  • 12 processing unit
  • 13 determination unit
  • 14 control unit
  • 15 storage unit
  • 2 extractor hood
  • 3 cooktop