Beverage preparation machine and method for operating a beverage preparation machine

Abstract

A beverage preparation machine and a method for operating the beverage preparation machine are provided. The beverage preparation machine comprises: a sensor of the first category, which provides only a single measurement value S1 at a specific time; a sensor of the second category, which provides a plurality of measurement values ​​S2 at a specific time; a data memory; and a control unit. The control unit is configured to compare the measurement value S2 with a target value stored in the data memory and, based on the comparison, trigger or not trigger at least one event. If the comparison does not lead to triggering an event, the detection of at least one measurement value S1 is achieved by the control unit, wherein the measurement value S1 is used to assign the measurement value S2 to at least one event, and these measurement values ​​S2 are stored in the data memory as target values ​​for the assigned event.

Description

[0001] A beverage preparation machine and a method for operating the beverage preparation machine are provided. The beverage preparation machine comprises: a sensor of the first category, which provides only a single measurement value S1 at a specific time; a sensor of the second category, which provides a plurality of measurement values ​​S2 at a specific time; a data memory; and a control unit. The control unit is configured to compare the measurement value S2 with a target value stored in the data memory and, based on the comparison, trigger or not trigger at least one event. If the comparison does not lead to the triggering of an event, the detection of at least one measurement value S1 is realized by the control unit, the measurement value S1 is used to assign the measurement value S2 to at least one event, and these measurement values ​​S2 are stored in the data memory as target values ​​for the assigned event.

[0002] Machines such as beverage preparation machines are primarily designed to accept commands from users and transform them into results. Thus, the commands (e.g. in the form of a whole chain of commands) are processed by the machine and completed into a result (desired by the user). Complex tasks can no longer be simply implemented by machines in this way, because it is very laborious to detect, define and program complex tasks with their boundary conditions to form a chain of commands. This process requires a lot of time, has a certain risk of failure and is associated with high costs.

[0003] In general language usage, with regard to computers, the term "artificial intelligence" is understood to mean that a computer is constructed or programmed so that it can solve problems autonomously. In the formation of an artificial neural network, it involves forming a network of nodes and connections on the software plane. The network is formed in software by deleting, adding or weighting connections or nodes.

[0004] Machine learning is based on algorithms and codes. In the case of machine learning, the machine acquires data, learns from the data, and makes decisions accordingly. The algorithm of machine learning can be represented as a flowchart or a decision tree. The branches of the flowchart or decision tree are weighted by the data collected by the machine.

[0005] Supervised learning of machines is aimed at a pre-established task to be learned, the results of which are known (status as of March 22, 2018: https: / / de.wikipedia.org / wiki / %C3%9Cberwachtes_Lernen). The results of the learning process can be compared with the known correct results, i.e. "monitored". In general, the learning steps are as follows:

[0006] 1. Application input;

[0007] 2. Process input (propagation);

[0008] 3. Compare the output with the expected value (error);

[0009] 4. Reduce the error by modifying the weights (e.g., using backpropagation).

[0010] In unsupervised learning by a machine, the target value is not known in advance. The machine tries to detect patterns and interpret those patterns.

[0011] In the case of backpropagation ("backpropagation" = error feedback), the algorithm operates in the following stages (status as of March 22, 2018: https: / / de.wikipedia.org / wiki / Backpropagation):

[0012] 1. The input pattern is applied and propagated forward through the network;

[0013] 2. Compare the network’s output to the expected output and take the difference between the two values ​​as the network’s error; and

[0014] 3. The error is now propagated back through the output to the input layer again, and the weights of the connections in the network are changed according to their effect on the error. This ensures that the output is close to the desired output when the input is reapplied.

[0015] A branch of machine learning is called deep learning. Here, large amounts of data are analyzed to detect and derive patterns from them.

[0016] Artificial intelligence is particularly advantageous if the machine is designed to implement a specific process based on complex detection patterns of sensors (e.g., the accumulation of multiple measurement points rather than on a single measurement value).

[0017] Sensors can basically be divided into two different categories, namely, Category 1 sensors and Category 2 sensors.

[0018] Sensors of category 1 provide a single measurement value only at a specific time, for example a specific temperature (temperature sensor), a specific pressure (pressure sensor), a specific speed (speed sensor), a specific flow rate (flow sensor) or a specific capacitance. If a sensor of this type measures permanently, a single measurement value is obtained as a function of time. The measurement results can be shown in an xy diagram, with x representing time and y representing the single measurement value. The example here can be a touch screen of a beverage preparation machine, which implements a specific event, such as outputting a specific beverage, when a specific pressure is exceeded (pressure-sensitive touch screen) or when a specific capacitance on a specific area of ​​the touch screen is detected (capacitive touch screen).

[0019] Sensors of the second category provide not only a single, but at least two different measured values ​​at a specific time, such as the frequency and amplitude of longitudinal oscillations (e.g. acoustic sensors or oscillation sensors), the frequency and amplitude of transverse oscillations (e.g. light sensors, IR sensors or radar sensors), the direction and strength of a magnetic field (magnetic field sensors) or the conductivity, pH value and sugar concentration of a specific liquid (taste sensors). If a sensor of this type measures permanently, at least two different measured values ​​are obtained as a function of time. In the case of two different measured values, the measurement result can be shown in an xyz diagram, with x representing the time, y representing the first measured value and z representing the second measured value.

[0020] Electrical devices (e.g. beverage preparation machines such as coffee machines) preferably operate with sensors of category 1, since with these sensors the measured values ​​to be determined are available and the calculations are uncomplicated. For example, a temperature sensor based on resistance outputs a specific resistance at a specific time, and conclusions about a specific temperature are drawn from this specific resistance.

[0021] The electrical device comprises software which can specify how the device should react when a specific measurement value is measured at a specific time. For example, when a specific threshold is exceeded, an action of the device can be triggered. Generally speaking, the software stores a table in which the measurement values ​​are assigned to specific device states or specific actions of the device (i.e. specific control commands for the device). In this way, the device itself can be controlled by a single measurement value of a sensor, for example a specific action can be implemented.

[0022] Electrical devices can also be controlled in principle using sensors of the second category. In this case, too, tables or patterns are applied, in which the assignment of specific measurement values ​​(patterns) detected by the sensor is coded to form one or more actions of the machine. However, in order to generate the assignment, the complexity is significantly higher than in the case of sensors of the first category (simpler). The reason for the higher complexity is that, for a command to cause the device to perform a specific action, at least two measurement values ​​must each take specific prescribed values ​​(the "and" operation of the y-value and the z-value at time x), thus increasing the complexity of the control.

[0023] It is assumed that in the future the increasing demands on devices (especially: beverage preparation machines such as coffee machines), namely complex control processes in the devices, can only be met with sensors from category 2.

[0024] It is therefore an object of the present invention to provide a beverage preparation machine, the complex control processes of which can be learned and implemented in a simple manner and with a minimum of complexity for the user.

[0025] The invention is achieved by a beverage preparation machine having the features of claim 1 and a method having the features of claim 8. The independent claims reveal advantageous developments.

[0026] According to the present invention, there is provided a beverage preparation machine comprising

[0027] a) at least one sensor of category 1, a sensor of category 1 being understood as a sensor which provides only a single measured value S1 at a given time;

[0028] b) at least one sensor of category 2, category 2 sensors being understood to be sensors which provide a plurality of measured values ​​S2 at a specific time;

[0029] c) data storage; and

[0030] d) a control unit configured to compare the measured value S2 with a target value of the measured value S2 stored in a data memory and coded for at least one specific event and to trigger or not trigger at least one event in the beverage preparation machine based on the comparison;

[0031] It is characterized in that the control unit of the beverage preparation machine is also configured to detect and use at least one measured value S1 in order to assign specific measured values ​​S2 that do not trigger an event after comparison with the target value of the measured value S2 to at least one event, and to store these measured values ​​S2 as target values ​​for at least one event in a data memory.

[0032] The beverage preparation machine according to the invention is therefore suitable for instructing one or more sensors of the second category (included in the beverage preparation machine) via one or more sensors of the first category (included in the beverage preparation machine) about specific, regularly occurring events. Here, an event is, for example, that the beverage preparation machine outputs espresso if at least one sensor of the second category detects an espresso cup at a specific location of the beverage preparation machine (e.g. under the brewing group), which is confirmed by the user via at least one sensor of the first category. Thus, the user's confirmation can be achieved by pressing a button, which confirmation can be achieved quickly and easily. The storage of the measured value S2 for this event in the data memory is automatically achieved.

[0033] An advantage of "teaching" the beverage preparation machine by means of at least one sensor of category 1 is that the complexity of the teaching is very low for a user of the beverage preparation machine (eg a simple pressing of a button is achieved).

[0034] If, after learning, the espresso cup is placed again in the detection area of ​​at least one sensor of the second category, the comparison of the pattern detected by the sensor of the second category with the pattern stored in the data memory, performed by the control unit of the beverage preparation machine, has the effect that the learned event is automatically triggered by the beverage preparation machine (i.e., without further input from the user). The learned automatic triggering event may involve, for example, displaying a "yes / no" query about a specific beverage (e.g., espresso), displaying a timer until a specific beverage (e.g., espresso) is output, displaying or highlighting a specific selection of beverages (e.g., espresso, ristretto, espresso macchiato), and / or immediately outputting espresso. If the learned event involves the beverage preparation machine immediately outputting espresso, espresso is therefore automatically output by the beverage preparation machine. In other words, the user is no longer required to reconfirm or select the output of espresso. If the learned event involves displaying or highlighting a specific selection of beverages, the specific selection of beverages is automatically displayed or highlighted by the beverage preparation machine. The user can then select a specific beverage from the selection. The selection can be made by sensors of the first category, for example by pressure on a button of the beverage preparation machine or by touching an area on a touch screen of the beverage preparation machine, or by sensors of the second category, for example by performing gestures and / or facial expressions (e.g. eye control) in front of a camera of the beverage preparation machine.

[0035] Thus, the beverage preparation machine according to the invention assigns the measured values ​​of the sensors of the second category to the events specified by the user based on the confirmation by the sensors of the first category, stores this assignment in its data memory and reacts accordingly in the future if the (known) event that has already been achieved occurs again. This assignment is achieved according to the principle of machine learning.

[0036] With the increase of the frequency of events during the operation of the beverage preparation machine, the beverage preparation machine is more experienced. Therefore, by exposing the beverage preparation machine to many events, the probability of a specific detection pattern being known on the sensor of the second category is increased, and therefore the probability of triggering an event without activating the sensor of the first category is also increased. In other words, the correlation between the pattern detected by the sensor of the second category and the associated event is increased. Accordingly, the importance of the sensor of the first category is gradually reduced, and no longer plays any role in the ideal case of the "taught" beverage preparation machine. The triggered event can also be, for example, in a specific form (i.e., different from before the event is triggered) to display the operating surface of the beverage preparation machine. Therefore, a specific beverage selection can be displayed or highlighted on the operating surface. The selection of a specific beverage in this beverage selection can be realized, for example, by pressing a button (sensor of the first category) or by a specific gesture (sensor of the second category).

[0037] In a preferred embodiment, the beverage preparation machine according to the invention comprises a user interface, preferably in the form of a touch screen. The user interface can be configured to display the triggered event. Furthermore, the user interface can be configured to display a countdown until the triggered event is achieved. Furthermore, the user interface can be configured to display a (specific) beverage selection. Furthermore, the user interface can be configured to display information about the measured value S2 of at least one sensor of the second category, preferably displaying information selected from the group consisting of the height of the beverage container, the level of the beverage container, the level of the bean container of the beverage preparation machine, the idling operation of the grinder of the beverage preparation machine, and a combination thereof. Furthermore, the user interface can be configured to display a target value for S2 and also to confirm and / or delete according to the user's wishes.

[0038] The control unit of the beverage preparation machine can be configured to trigger at least one event, which is encoded in the target value of the measured value S2 of the sensor of the second category, if the measured value S2 can be assigned to the target value of the measured value S2 with a probability of >50%, preferably ≥60%, particularly preferably ≥70%, very particularly preferably ≥80%, and in particular ≥90%.

[0039] In a preferred embodiment, the control unit of the beverage preparation machine is configured to use at least one measured value S1 if the measured value S2 can be assigned to a target value of the measured value S2 with a probability of only ≤50%, preferably ≤40%, particularly preferably ≤30%, very particularly preferably ≤20%, in particular ≤10%, respectively.

[0040] At least one sensor of category 1 may be configured to provide only one-dimensional measurement S1 at a particular time. The one-dimensional measurement may be selected from the group consisting of temperature, pressure, velocity, flow rate, capacitance, direction of movement, distance, time and mass.

[0041] The at least one sensor of the first category can be selected from the group consisting of a temperature sensor, a pressure sensor, a speed sensor, a flow rate sensor, a current sensor and a capacitive sensor. Preferably, the at least one sensor of the first category is a pressure sensor and / or a capacitive sensor. Very particularly preferably, the sensor of the first category is a touch screen of the beverage preparation machine.

[0042] At least one sensor of the second category can be configured to provide a multi-dimensional measurement value S2 at a specific time, preferably a measurement value S2 of at least three dimensions, and particularly preferably a measurement value S2 of at least four dimensions. The dimension of the multi-dimensional measurement value S2 can be selected from the group consisting of oscillation frequency (e.g., light wavelength, i.e., light color), oscillation amplitude (e.g., light amplitude, i.e., light intensity), oscillation polarization (e.g., light polarization), oscillation propagation direction (e.g., light propagation direction), force line propagation direction (e.g., magnetic field line propagation direction), and force line intensity (e.g., magnetic field line intensity). Similarly, another dimension of the multi-dimensional measurement value S2, i.e., another dimension different from the (first) dimension, can be selected from the above group.

[0043] At least one sensor of the second category may be selected from the group consisting of sensors for detecting longitudinal oscillations, sensors for detecting transverse oscillations, sensors for detecting magnetic fields, sensors for detecting electric fields, chemical sensors, electrical sensors and combinations thereof. Preferably, the sensors of the second category are selected from the group consisting of sound sensors, vibration sensors, optical sensors (e.g. cameras), IR sensors, radar sensors, magnetic field sensors, conductivity sensors, pH sensors, sensors for detecting sugar concentration and combinations thereof. Preferably, the sensors of the second category are radar sensors, optical sensors (e.g. line cameras or matrix cameras), sound sensors (e.g. acoustic sensors) and combinations thereof.

[0044] Particularly preferably, the radar sensors are sensors of category 2. They have the advantage that they provide very precise information about the properties of a specific beverage container and thus enable a fine and precise differentiation between different beverage containers.

[0045] Optical sensors, such as line cameras or matrix cameras, are likewise particularly preferred sensors of category 2. They provide, at time x in a matrix y / z, a value y relating to, for example, a specific light intensity (light brightness) and a value z relating to, for example, a specific light frequency (light color). In the case of a ToF sensor, the spatial distance relative to the sensor can be detected as an additional dimension. Optical sensors such as radar sensors are suitable for providing information about the properties of a specific beverage container.

[0046] A sound sensor (e.g. an acoustic sensor) is also a particularly preferred sensor of the second category. The sound sensor is suitable, for example, for detecting the grinding noise of a grinder (e.g. for coffee beans) of a beverage preparation machine and preventing the grinder from running empty. To achieve this function, the sound sensor is preferably mounted near the grinder of the beverage preparation machine. The advantage of using such a sound sensor as a sensor of the second category is that, by means of the sound curve recorded by the sensor, the sound pattern can be used shortly before the grinder runs empty to indicate to the user that the grinder is about to run empty. By predicting that the grinder is about to run empty, the user can be prompted to refill the bean container and thus the emptying of the bean container during the dispensing of a beverage (e.g. coffee) can be prevented. This has the effect that the provision of a beverage (e.g. coffee) can be achieved without interruption and without downtime in the beverage preparation machine. In addition, it is a known problem that the grinder, once running empty, requires a certain lead time (a certain grinding time) in order to fill the bean container with (ground) beans again after it has been filled. On the one hand, this prolongs the time until the beverage is provided after the beverage dispensing is triggered. On the other hand, the result is a certain variability in the metering relative to the amount of ground beans used to prepare the beverage and therefore relative to the composition of the beverage dispensed after the grinder has run empty. The use of a sound sensor has the effect that the grinder can be prevented from running empty right from the start. Thus, the beverage preparation machine can always provide a uniform product composition for the beverage without downtime and without delays during dispensing.

[0047] In a preferred embodiment, at least one specific event (which can be triggered within the beverage preparation machine) is selected from the group consisting of:

[0048] i) interaction of the beverage preparation machine with the user, preferably selected from the group consisting of outputting a request of the beverage preparation machine to the user, displaying a specific beverage selection, fading out a specific beverage selection and combinations thereof, the request being particularly preferably selected from the group consisting of refilling coffee beans, emptying the salt container, filling the water tank, filling the milk tank, starting a care program (e.g. starting a cleaning program, starting a descaling program and / or starting a cleaning program) and combinations thereof;

[0049] ii) maintaining the beverage preparation machine, preferably selected from the group consisting of detecting and / or evaluating a fault in the operation of the beverage preparation machine, detecting and / or evaluating a fault in a component of the beverage preparation machine, determining a service interval for the beverage preparation machine, triggering a service call for the beverage preparation machine, setting and / or adjusting adjustments of the beverage preparation machine;

[0050] iii) controlling at least one component of the beverage preparation machine, preferably outputting a specific beverage through the beverage preparation machine, moving a beverage outlet of the beverage preparation machine, implementing a care program of the beverage preparation machine and combinations thereof; and

[0051] iv) combinations thereof.

[0052] Furthermore, according to the invention there is provided a method for operating a beverage preparation machine comprising the following steps

[0053] a) detecting a plurality of measured values ​​S2 from at least one sensor of the second category, a sensor of the second category being understood to be a sensor which provides a plurality of measured values ​​S2 at a certain time;

[0054] b) comparing, by a control unit of the beverage preparation machine, the measured value S2 with a target value for the measured value S2, which target value is stored in a data memory of the beverage preparation machine and is coded for at least one specific event;

[0055] c) triggering or not triggering at least one event in the beverage preparation machine based on the comparison; and

[0056] d) if no event is triggered, detecting at least one measured value S1 from at least one sensor of category 1, a sensor of category 1 being understood as a sensor which provides only a single measured value S1 at a specific time;

[0057] It is characterized in that, when no event is triggered after comparison with a target value of the measured value S2, the control unit uses at least one measured value S1 from at least one sensor of the first category of the beverage preparation machine to assign at least one event to the measured value S2 that did not trigger an event after comparison with the target value of the measured value S2, and these measured values ​​S2 are stored in the data storage device as target values ​​for at least one event.

[0058] The beverage preparation machine used in the method according to the invention may comprise a user interface, preferably a touch screen. Preferably, the triggered event is displayed via the user interface. Furthermore, a countdown until the realization of the triggered event can be displayed via the user interface. Furthermore, via the user interface, a (specific) beverage selection can be displayed. Furthermore, information about the measured value S2 of at least one sensor of the second category can be displayed via the user interface, preferably information selected from the group consisting of the height of the beverage container, the level of the beverage container, the level of the bean container of the beverage preparation machine, the idling operation of the grinder of the beverage preparation machine, and combinations thereof. Furthermore, the target value of S2 can be displayed via the user interface and can be confirmed and / or deleted according to the user's wishes.

[0059] In a preferred embodiment, if the measured value S2 can be assigned to the target value of the measured value S2 with a probability of >50%, preferably ≥60%, particularly preferably ≥70%, very particularly preferably ≥80%, and in particular ≥90%, at least one event encoded in the target value of the measured value S2 is triggered.

[0060] In the method according to the invention, for example, at least one measured value S1 can be detected and used if the measured value S2 can be assigned to a target value of the measured value S2 with a probability of only ≤50%, preferably ≤40%, particularly preferably ≤30%, very particularly preferably ≤20%, in particular ≤10%.

[0061] In a preferred embodiment of the method, at least one sensor of category 1 provides a one-dimensional measurement value S1 at a specific time. The one-dimensional measurement value can be selected from the group consisting of temperature, pressure, speed, flow rate, capacitance, direction of movement, distance, time and mass.

[0062] The sensor of category 1 used in the method may have the same properties as the above-mentioned sensor of category 1 of the beverage preparation machine.

[0063] Furthermore, it is preferred that at least one sensor of the second category is used in the method, which sensor provides multidimensional measured values ​​S2 at a specific time, preferably at least three-dimensional measured values ​​S2, in particular at least four-dimensional measured values ​​S2. The dimensions of the multidimensional measured values ​​can be selected from the group consisting of oscillation frequency (e.g. light wavelength, i.e. light color), oscillation amplitude (e.g. light amplitude, i.e. light intensity), oscillation polarization (e.g. light polarization), oscillation propagation direction (e.g. light propagation direction), force line propagation direction (e.g. magnetic field line propagation direction), force line strength (e.g. magnetic field line strength). Other dimensions of the multidimensional measured values ​​can also be selected from the above group.

[0064] The sensor of category 2 used in the method may have the same properties as the above-mentioned sensor of category 2 of the beverage preparation machine.

[0065] The method is characterized in that the at least one specific event is selected from the group consisting of:

[0066] i) interaction of the beverage preparation machine with the user, preferably selected from the group consisting of outputting a request of the beverage preparation machine to the user, displaying a specific beverage selection, fading out a specific beverage selection and combinations thereof, the request being particularly preferably selected from the group consisting of refilling coffee beans, emptying the salt container, filling the water tank, filling the milk tank, starting a care program and combinations thereof;

[0067] ii) maintaining the beverage preparation machine, preferably selected from the group consisting of detecting and / or evaluating a fault in the operation of the beverage preparation machine, detecting and / or evaluating a fault in a component of the beverage preparation machine, determining a service interval for the beverage preparation machine, triggering a service call for the beverage preparation machine, setting and / or adjusting adjustments of the beverage preparation machine;

[0068] iii) controlling at least one component of the beverage preparation machine, preferably outputting a specific beverage through the beverage preparation machine, moving a beverage outlet of the beverage preparation machine, implementing a care program of the beverage preparation machine and combinations thereof; and

[0069] iv) combinations thereof.

[0070] The subject matter according to the invention will be explained in more detail with reference to the following figures and examples, without wishing to restrict the latter to the specific embodiments represented herein.

[0071] Figure 1 The functional mechanism of a beverage preparation machine according to the invention is shown. If the object to be analyzed is detected by a sensor 2 of the second category, a plurality of measured values ​​4 (so-called patterns) are sent from the sensor 2 of the second category to a control unit 6 of the beverage preparation machine. The control unit 6 accesses a data memory 5 and checks whether the detected measured values ​​4 are already known with a certain probability, i.e. whether these measured values ​​are stored in the data memory 5 and assigned there to a specific event. If this question 7 can be answered in the affirmative, the beverage preparation machine triggers a specific event 8. If the question is not answered in the affirmative, the beverage preparation machine waits for a single measured value 3 of a sensor 1 of the first category. After obtaining the measured values ​​3 of the sensor 1 of the first category, the measurement values ​​4 detected by the sensor 2 of the second category are assigned to at least one specific event 8 in the beverage preparation machine. In the data memory 5, initially no pattern 4 is associated with an event 8. The measured values ​​3 of the sensor of the first category are then used to associate the measured values ​​4 of the two measurements of the sensor of the second category with a specific event 8, and this association is stored in the data memory 5 (= machine learning).

[0072] Figure 2 The functional arrangement of a beverage preparation machine according to the invention is shown, which has a radar sensor 2 for detecting a plurality of measured values ​​12 , 12 ′ of a container 10 , 10 ′ placed by a user below a beverage dispensing unit 11 of the beverage preparation machine.

[0073] exist Figure 2 In A, the container is a latte macchiato glass 10, which generates a spectrum 12 from the radar sensor 2 that is characteristic of the glass 10. Figure 2In case A, the characteristic spectrum 12 is already stored in the data memory 5 of the beverage preparation machine (see √ symbol), so that the beverage preparation machine recognizes the latte-macchiato glass 10 and implements an event associated with the glass 10 (e.g. output of the latte-macchiato glass by the beverage output unit 11 of the beverage preparation machine or display or highlighting of a specific beverage selection) without further input from the user (e.g. by interaction with the sensor 1 of the first category). If the event is display or highlighting of a specific beverage selection (e.g. an enlarged display of a specific beverage on the user face of the beverage preparation machine), the selection and output of a specific beverage from the beverage selection can be implemented specifically by the user using pressure on a button (activation of a sensor of the first category) or a gesture, or a facial expression or a voice speech (activation of a sensor of the second category).

[0074] The same applies to Figure 2 The situation shown in B. Figure 2 In B, an espresso cup 10' is placed under the beverage output unit 11 of the beverage preparation machine instead of the latte-macchiato glass 10. Since the characteristic spectrum 12' of the espresso cup 10' detected by the radar sensor 2 has been stored in the data memory 5 (see the √ symbol), the beverage preparation machine recognizes the espresso cup 10' itself and, without further input from the user (e.g., by interaction with the sensor 1 of the first category), executes the event associated with the cup 10' (e.g., outputting espresso through the beverage output unit 11 of the beverage preparation machine).

[0075] exist Figure 2 C, on the other hand, shows a situation in which the pattern 12 for the latte macchiato glass 10 detected by the radar sensor 2 is not yet stored in the data memory 5 of the beverage preparation machine or has no similarity to a pattern stored there (see X symbol). This has the effect that the beverage preparation machine uses the measured value detected by the sensor 1 of category 1 (here the touch screen 1 on the beverage preparation machine) (here the impedance change caused by the user touching the latte macchiato glass 10 displayed on the touch screen) to assign the detected pattern 12 to the event "output of latte macchiato" and stores this assignment in the data memory 5.

[0076] Figure 3 An example of a number of measurements of sensors of category 2 is shown. Sensors of category 2 are two-dimensional sensors which detect the intensity (y-axis = 1st dimension) of electromagnetic radiation of different frequencies I, II, III (z-axis = 2nd dimension) over time (x-axis).

[0077] Example 1 - Beverage preparation machine with electromagnetic sensor for cup recognition

[0078] The beverage preparation machine is a coffee machine, in which a sensor from category 2 is integrated. In this case, the sensor is a radar sensor (for example a 60 GHz radar sensor from Infineon Technologies AG with sensitivity in the submillimeter range) or an image sensor.

[0079] First, a glass or cup is placed under the beverage preparation machine. In the case of a radar sensor, the frequency of the detected radar radiation is measured as a first measurement value, and the amplitude of the detected radar radiation is measured as a second measurement value. In the case of an image sensor, the frequency of the detected light (color of the light) and the amplitude of the detected light (brightness of the light) are detected.

[0080] Since no target values ​​are defined in the permanent memory of the beverage preparation machine, no measured values ​​(ie patterns) are assigned to specific events either.

[0081] The beverage preparation machine has a user input interface, such as a touch pad, as a sensor of category 1. Via the user input interface, various beverages that can be delivered by the beverage preparation machine are displayed to the user.

[0082] If the user has placed a latte macchiato glass under the beverage preparation machine because he wants to make a latte macchiato, the user now selects the output of latte macchiato by pressure on the touchpad. The signal of the sensor of category 1 (touchpad) is now used to associate the signal recorded by the radar sensor or image sensor with the output of latte macchiato and store this association in a permanent data memory. If the same latte macchiato glass is placed under the beverage preparation machine the next time, the output of latte macchiato is achieved without activation by the touchpad, because the beverage preparation machine has already assigned the pattern of the specific latte macchiato glass detected by the sensor of category 2 to the output of the latte macchiato glass.

[0083] Since the shape of the latte-macchiato glass can vary within a series or between different manufacturers, the more often the latte-macchiato glass is placed under the beverage preparation machine (shape information increases), the better the allocation becomes. In addition, when a specific glass is reused, the accuracy is improved because the sensors of the second category will also have measurement errors, the importance of which decreases with each further repetition of the measurement. At some time or another, user input through the touchpad is no longer required because the machine explicitly assigns the detected measurement value to the latte-macchiato glass. On the user surface (e.g., display or touch screen) of the beverage preparation machine, as a result, only beverages can be displayed, which have matched the last output with the same mode S2 and the button of appropriately pressing S1 based on a comparison with the data storage device (here: latte-macchiato). Then, the user has the choice of beverage (here: only latte-macchiato is displayed on the user surface) or may cancel the selection of beverage with another command.

[0084] Of course, the control unit of the beverage preparation machine can be configured so that the beverage preparation machine offers the user a choice of beverages and / or blocks a specific beverage after the container has been placed down. If no alternative beverage is available for the latte-macchiato glass, the control unit can be configured so that the output of the latte-macchiato starts automatically. This can be achieved, for example, by displaying "Beverage starts automatically in X seconds. To cancel, press Y", X representing seconds and Y representing a button to be pressed on the beverage preparation machine.

[0085] For example, a radar sensor is used as a sensor of category 2 so that part of the radar wave is reflected on the surface of the object to be inspected (e.g. a coffee cup) and another part penetrates before being reflected into the object, slowed down due to the higher density of the object compared to the surrounding air, and then comes out of the object again. The number of reflected beams and the difference in the run time between the reflected beam and the transmitted beam are properties of different objects and materials.

[0086] Furthermore, further information about the glass or cup can be collected by additional sensors in the beverage preparation machine and can be assigned to the sensor signal. The following sensors are possible:

[0087] - Collision detection sensor to determine the height of the container;

[0088] - Flow sensors to prevent containers from overflowing;

[0089] - A pressure sensor for cancelling the dispensing of the beverage so as to adapt the volume to the normal usage of the user.

[0090] Example 2 - Beverage preparation machine with acoustic sensor for beans-empty reporting

[0091] The beverage preparation machine is a coffee machine and comprises an acoustic sensor as a sensor of category 2. The acoustic sensor "hears" various events or errors. For example, when running empty, the grinder of the coffee machine emits a characteristic noise. This noise can be recorded by the acoustic sensor and associated with an event that the grinder should stop the grinding process. If after this "learning process" the acoustic sensor detects the characteristic noise again, the grinder can be stopped before the result is that the grinder runs completely empty.

[0092] In the prior art, for example, only one current sensor provides information about whether the grinder motor is running idling (i.e. no longer grinding coffee beans). In this case, the level of the current detected by the current sensor drops. Now, the noise pattern emitted by the grinder can be detected and learned, for example, actually 2 to 3 seconds before the current limit value is reached. By this measure, the beverage preparation machine detects the state shortly before the grinder "becomes empty" and can turn off the grinder in good time (before a lower current value is detected). The advantage is that the product preparation is not interrupted by refilling beans: when the grinder is running idling, the grinder and the grinder outflow are empty, and after filling with coffee beans, a specific lead time (grinding time) is required in order to fill the system with beans and ground coffee. Only after this, the coffee grounds are correctly transported to the brewing chamber again. Therefore, a time loss is generated, resulting in a metering change with respect to the ground coffee grounds.

[0093] Likewise, noise during refilling of coffee beans can be detected as well as specific events associated with this process.

[0094] Reference numerals list

[0095] 1: Class 1 sensors (e.g. touch displays);

[0096] 2: Category 2 sensors (e.g. radar sensors);

[0097] 3: Individual measurement values ​​S1 of sensors of category 1 (e.g. discrete pressure values ​​of a pressure sensor);

[0098] 4: a plurality of measurements S2 of sensors of category 2 (e.g. with locally resolved discrete frequencies and discrete intensities of electromagnetic radiation);

[0099] 5: Data storage;

[0100] 6: Control unit;

[0101] 7: Determine whether the measured value 4 exists in the data memory 5;

[0102] 8: Event triggered by the beverage preparation machine;

[0103] 9: Assign measurement value 4 to at least one event;

[0104] 10: Latte-Macchiato glass;

[0105] 10': Espresso cup;

[0106] 11: Beverage output unit of the beverage preparation machine;

[0107] 12: spectrum of latte-macchiato glass 10;

[0108] 12': spectrum of a 10' espresso cup;

[0109] I: The first frequency of electromagnetic radiation;

[0110] II: The second frequency of electromagnetic radiation;

[0111] III The 3rd frequency of electromagnetic radiation.

Claims

1. A beverage preparation machine, comprising a) at least one sensor of category 1, wherein a sensor of category 1 is understood to be a sensor that provides only one single measurement value S1 at a specific time, wherein the sensor of category 1 is selected from the group consisting of a temperature sensor, a velocity sensor, a flow rate sensor, a current sensor, a capacitive sensor and a user input interface; b) at least one sensor of the second category, which is understood to be a sensor that provides a plurality of measured values ​​S2 at a certain time, and which is selected from the group consisting of sound sensors, vibration sensors, radar sensors, magnetic field sensors, conductivity sensors, pH sensors, sensors for detecting sugar concentration, and combinations thereof; c) data storage; d) a control unit configured to compare the measured value S2 with a target value of the measured value S2 stored in the data memory and coded for at least one specific event and to trigger or not trigger at least one event in the beverage preparation machine based on the comparison; It is characterized in that the control unit of the beverage preparation machine is also configured to detect and use at least one measurement value S1 so as to assign a specific measurement value S2 that does not trigger an event after comparison with a target value of the measurement value S2 to at least one event, and to store the specific measurement value S2 as the target value of the at least one event in the data storage device.

2. The beverage preparation machine according to claim 1, characterized in that The beverage preparation machine comprises a user interface configured to i) display the triggered event; and / or ii) Display a countdown until the triggered event is achieved; and / or iii) displaying beverage selections; iv) displaying information about a measurement value S2 of at least one sensor of the second category; and / or v) Display the target value of S2 and confirm or delete it according to the user's wishes.

3. The beverage preparation machine according to claim 2, characterized in that The user interface includes a touch screen.

4. The beverage preparation machine according to claim 2, characterized in that The information about the measured value S2 of at least one sensor of the second category comprises information selected from the group consisting of the height of the beverage container, the level of the beverage container, the level of the bean container of the beverage preparation machine, the idling operation of the grinder of the beverage preparation machine and combinations thereof.

5. The beverage preparation machine according to claim 1 or 2, characterized in that: The control unit is configured to trigger at least one event encoded in the target value of the measured value S2 if the measured value S2 is respectively assigned to the target value of the measured value S2 with a probability of >50%.

6. The beverage preparation machine according to claim 5, characterized in that The control unit is configured to trigger at least one event encoded in the target value of the measured value S2 if the measured value S2 is respectively assigned to the target value of the measured value S2 with a probability of ≧60%.

7. The beverage preparation machine according to claim 5, characterized in that The control unit is configured to trigger at least one event encoded in the target value of the measured value S2 if the measured value S2 is respectively assigned to the target value of the measured value S2 with a probability of ≧70%.

8. The beverage preparation machine according to claim 5, characterized in that The control unit is configured to trigger at least one event encoded in the target value of the measured value S2 if the measured value S2 is respectively assigned to the target value of the measured value S2 with a probability of ≧80%.

9. The beverage preparation machine according to claim 5, characterized in that The control unit is configured to trigger at least one event encoded in the target value of the measured value S2 if the measured value S2 is respectively assigned to the target value of the measured value S2 with a probability of ≥90%.

10. The beverage preparation machine according to claim 1 or 2, characterized in that The control unit is configured to use at least one measured value S1 if the measured value S2 is respectively assigned to a target value for the measured value S2 with only a probability of ≤50%.

11. The beverage preparation machine according to claim 10, characterized in that The control unit is configured to use at least one measured value S1 if the measured value S2 is respectively assigned to a target value for the measured value S2 with a probability of only ≤40%.

12. The beverage preparation machine according to claim 10, characterized in that The control unit is configured to use at least one measured value S1 if the measured value S2 is respectively assigned to a target value for the measured value S2 with a probability of only ≤30%.

13. The beverage preparation machine according to claim 10, characterized in that The control unit is configured to use at least one measured value S1 if the measured value S2 is respectively assigned to a target value for the measured value S2 with a probability of only ≤20%.

14. The beverage preparation machine according to claim 10, characterized in that The control unit is configured to use at least one measured value S1 if the measured value S2 is respectively assigned to a target value for the measured value S2 with a probability of only ≤10%.

15. The beverage preparation machine according to claim 1 or 2, characterized in that The at least one sensor of category 1 is configured to provide a measurement value S1 in only one dimension at a specific time.

16. The beverage preparation machine according to claim 1 or 2, characterized in that The at least one sensor of the second category is configured to provide multi-dimensional measurement values ​​S2 at a specific time.

17. The beverage preparation machine according to claim 16, characterized in that The at least one sensor of the second category is configured to provide at least three-dimensional measurement values ​​S2 at a specific time.

18. The beverage preparation machine according to claim 16, characterized in that The at least one sensor of the second category is configured to provide measurement values ​​S2 in at least four dimensions at a specific time.

19. The beverage preparation machine according to claim 1 or 2, characterized in that The at least one specific event is selected from the group consisting of interaction of the beverage preparation machine with a user, maintenance of the beverage preparation machine, control of at least one component of the beverage preparation machine, and combinations thereof.

20. A method for operating a beverage preparation machine comprising the steps of a) detecting a plurality of measured values ​​S2 from at least one sensor of the second category, wherein a sensor of the second category is understood to be a sensor that provides a plurality of measured values ​​S2 at a certain time, the sensor of the second category being selected from the group consisting of a sound sensor, a vibration sensor, a radar sensor, a magnetic field sensor, a conductivity sensor, a pH sensor, a sensor for detecting sugar concentration, and combinations thereof; b) comparing, by a control unit of the beverage preparation machine, the measured value S2 with a target value for the measured value S2, the target value being stored in a data memory of the beverage preparation machine and being coded for at least one specific event; c) triggering or not triggering at least one event in the beverage preparation machine based on the comparison; as well as d) if no event is triggered, detecting at least one measurement value S1 from at least one sensor of category 1, category 1 sensors being understood as sensors that provide only one single measurement value S1 at a specific time, wherein said category 1 sensors are selected from the group consisting of temperature sensors, speed sensors, flow rate sensors, current sensors, capacitive sensors and user input interfaces; It is characterized in that, when no event is triggered after comparison with the target value of the measurement value S2, the control unit uses at least one measurement value S1 from at least one sensor of the first category of the beverage preparation machine to assign at least one event to a specific measurement value S2 that does not trigger an event after comparison with the target value of the measurement value S2, and the specific measurement value S2 is stored in the data storage device as the target value of the at least one event.

21. The method according to claim 20, characterized in that The beverage preparation machine comprises a user interface through which i) display the triggered event; and / or ii) Display a countdown until the triggered event is achieved; and / or iii) displaying beverage selections; iv) displaying information about a measurement value S2 of at least one sensor of the second category; and / or v) Display the target value of S2 and confirm or delete it according to the user's wishes.

22. The method according to claim 21, characterized in that The user interface includes a touch screen.

23. The method according to claim 21, characterized in that The information about the measured value S2 of at least one sensor of the second category comprises information selected from the group consisting of the height of the beverage container, the level of the beverage container, the level of the bean container of the beverage preparation machine, the idling operation of the grinder of the beverage preparation machine and combinations thereof.

24. The method according to claim 20 or 21, characterized in that If a measured value S2 is respectively assigned to a target value of the measured value S2 with a probability of >50%, at least one event encoded in the target value of the measured value S2 is triggered.

25. The method according to claim 24, characterized in that If measured value S2 is respectively assigned to a target value for measured value S2 with a probability of ≥60%, at least one event encoded in the target value for measured value S2 is triggered.

26. The method according to claim 24, characterized in that If measured value S2 is respectively assigned to a target value for measured value S2 with a probability of ≥70%, at least one event encoded in the target value for measured value S2 is triggered.

27. The method according to claim 24, characterized in that If measured value S2 is respectively assigned to a target value for measured value S2 with a probability of ≥80%, at least one event encoded in the target value for measured value S2 is triggered.

28. The method according to claim 24, characterized in that If measured value S2 is respectively assigned to a target value for measured value S2 with a probability of ≥90%, at least one event encoded in the target value for measured value S2 is triggered.

29. The method according to claim 20 or 21, characterized in that At least one measured value S1 is detected and used if the measured values ​​S2 are respectively assigned to a target value for the measured value S2 with a probability of only ≤50%.

30. The method according to claim 29, characterized in that If the measured values ​​S2 are respectively assigned to a target value for the measured value S2 with only a probability of ≤40%, at least one measured value S1 is used.

31. The method according to claim 29, characterized in that If the measured values ​​S2 are respectively assigned to a target value for the measured value S2 with only a probability of ≤30%, at least one measured value S1 is used.

32. The method according to claim 29, characterized in that At least one measured value S1 is used if the measured values ​​S2 are respectively assigned to a target value for the measured value S2 with a probability of only ≤20%.

33. The method according to claim 29, characterized in that At least one measured value S1 is used if the measured values ​​S2 are respectively assigned to a target value for the measured value S2 with a probability of only ≤10%.

34. The method according to claim 20 or 21, characterized in that The at least one sensor of the first category supplies a one-dimensional measured value S1 at a specific time.

35. The method according to claim 20 or 21, characterized in that The at least one sensor of the second category provides multi-dimensional measured values ​​S2 at a specific time.

36. The method according to claim 35, characterized in that The at least one sensor of the second category provides at least three-dimensional measured values ​​S2 at a specific time.

37. The method according to claim 35, characterized in that The at least one sensor of the second category provides at least four-dimensional measured values ​​S2 at a specific time.

38. The method according to claim 20 or 21, characterized in that The at least one specific event is selected from the group consisting of interaction of the beverage preparation machine with a user, maintenance of the beverage preparation machine, control of at least one component of the beverage preparation machine, and combinations thereof.

Images