Patent Application
20190306259
The present invention relates to a method of operating and/or controlling a personal care device, in particular a hair removal device such as an electric shaver, wherein at least one user's behavior parameter characterizing the user's behavior when handling the personal care device is detected by at least one detector during the personal care treatment. Furthermore, the present invention also relates to such personal care device comprising a working head for effecting a personal care treatment to a body surface, a handle for moving the working head along the body surface, and at least one detector for detecting at least one user's behavior parameter characterizing the user's behavior when handling the personal care device during a personal care treatment session.
Personal care devices are widely used to apply different types of personal care treatments to users, wherein such personal care devices may include hair removal devices such as epilators, shavers or razors which may be electric or manual and/or wet or dry, or beard trimmers. Furthermore, other personal care devices include dental care appliances such as electric or manual toothbrushes, interdental cleaners or gum massaging devices, or skin treatment devices such as massaging devices or vibrators. All such personal care devices are subject to the problem that different people behave in different ways and thus, different users use the personal care device in different ways. Even when the personal care device provides for self-adjustment of the setting of some of its working parameters, it is difficult to fit different users' characteristics.
In a more general context, different people behave in different ways. For example, when cleaning the same table with the same amount of dirt with the same cloth, some people will press hard and others will press more lightly. Some will use circling motions, while others will move the cloth in straight lines. These different behaviors come naturally to the different people. If Person A would be forced to use the cloth in the way Person B does and vice versa, then they would find this type of usage to require more effort (physical and/or concentration).
A perfectly developed product offers an easy and intuitive usage for every user. In particular, there is no need for the user to be highly concentrated, make unergonomical movements or exert extra effort during the usage of the product—in other words the user can operate the product using his/her natural behavior and does not need to alter this behavior as otherwise the product does not work well enough.
For complex products (e.g. products with more than one component, mechanical products, electrical/electronic products, etc.), a product design that performs well with a multitude of different natural behaviors is however very difficult to achieve, in particular as humans behave in very many different ways, thus no single product design will perform optimally for all natural behaviors of all people. One possible solution to this is to design a product that can adapt to fit the different natural behaviors of different users.
In order to do this however, the product needs know when and how to adapt, in other words it is necessary to identify when the user deviates from his/her natural behavior and what his/her natural behavior is. This is not a simple task as people behave in very different ways and what might be natural behavior for one person may not be a natural behavior for another. Also, it is advantageous to understand what the user is trying to achieve with his non-natural behavior (i.e. what the product is not delivering when the person only uses his natural behavior) in order for the product to adapt appropriately.
For example, electric shavers usually have one or more cutter elements driven by an electric drive unit in an oscillating manner where the cutter elements reciprocate under a shearfoil, wherein such cutter elements or undercutters may have an elongated shape and may reciprocate along their longitudinal axis. Other types of electric shavers use rotatory cutter elements which may be driven in an oscillating or a continuous manner Said electric drive unit may include an electric motor or an electric-type linear motor, wherein the drive unit may include a drive train having elements such as an elongated drive transmitter for transmitting the driving motion of the motor to the cutter element, wherein the motor may be received within the handle portion of the shaver or in the alternative, in the shaver head thereof.
Although such personal care devices such as shavers are used on a daily basis by most users, it is sometimes difficult to operate and handle the shaver indeed perfectly. Due to different preferences and habits of different users, the shaver is not operated in its optimum range even when the shaver is adjustable. For example, the working head with the cutter elements may be pressed against the skin too strongly, or the shaver may be held at an orientation preventing the working head's shear foils from full contact with the skin, even if the working head is pivotably supported to compensate for some angular displacement. Sometimes it is also difficult to move the shaver along the skin at the right velocity in the right direction to the relevant skin portions.
Document EP 0 720 523 B1 discloses an electric shaving apparatus which allows for adjusting the height over which the cutter elements project from the shaver head surface, adjusting the pretensioning force of the cutter blades against which pretensioning force the cutter blades may dive, and adjusting the motor speed so as to balance shaving performance and skin irritation. Said adjustable parameters are automatically controlled in response to a plurality of detected working parameters by means of fuzzy logic to balance the influence of the different input signals indicative of the different working parameters.
Furthermore, WO 2007/033729 A1 discloses an electric hair removal device adjusting the motor speed and thus cutter speed in response to the velocity at which the hair removal device is moved along the user's skin which velocity is measured by means of a rotational sensor. The shaver includes a memory in which velocity detected in the past is stored so as to start a hair removal session with a motor speed in line with the stored velocity detected in the past.
Document WO 2015/067498 A1 discloses a hair cutting device, wherein a position identifier including cameras identifies the position of the hair cutter relative to the body part to be treated, wherein a feedback module gives feedback to indicate the desired path and the desired angle of orientation of the cutter relative to the body part.
Furthermore, document WO 2017/062326 A1 describes a personal care device linked to a smartphone and a computer system via a network so as to monitor device usage. More particularly, working time is monitored to indicate when a replacement part such as a razor cartridge needs to be replaced, wherein determination of working time includes adjustment of the sensor settings such as the minimum duration for counting a shaver stroke.
Furthermore, document WO 2017/032547 A1 discloses a shaving device giving a user shaving instructions acoustically and/or visually, wherein such shaving instructions such as “user gentle pressure only” or “use sensitive speed setting” are given based on usage data such as pressure data and/or motion data measured by the shaving device. It is also suggested to take into account usage data history to select the appropriate instruction from a stored list of instructions. The user has to manually input some personal assessment data which may comprise information relating to skin sensitivity, age, hair type, uneven skin, in-growing hair, difficulties to catch hair, etc., which the user may provide via a user interface in response to a questionnaire presented via the user interface.
EP 1549468 B1 describes a shaver which detects proper contact of the shear foils with the skin to be shaved, wherein it is mentioned that such contact may be detected by means of an inductive sensor, a capacitance sensor or an optical sensor which may include a light barrier immediately above the shear foil. It is suggested to automatically vary the position of the shaver head relative to the handle by means of an actuator for pivoting or tilting the shaver head, when there is improper contact to the skin.
It is an objective underlying the present invention to provide for an improved personal care device and an improved method of operation thereof, avoiding at least one of the disadvantages of the prior art and/or further developing the existing solutions. A more particular objective underlying the invention is to provide for an improved method and personal care device at least reducing non-natural behavior of the user including adaption of the user to the personal care device.
A further objective underlying the invention is to provide for an improved personal care device requiring less adaption from the user to the product.
A still further objective underlying the invention is to provide for an improved method of operating and/or controlling a personal care device to achieve better fit to different behavior and preferences of different users.
To achieve at least one of the aforementioned objectives, it is suggested to compare the real-time data of the at least one behavioral parameter as detected by the at least one detector when handling the personal care device during a personal care treatment to historical data of at least one behavioral parameter as stored in a storage and/or to detect at least one emotional parameter by means of an emotion detector indicative of the user's emotion when handling the personal care device during a personal care treatment session, and to determine from said comparison of the real-time data of the user's behavior to the historical data thereof and/or to the detected emotional parameter whether the user's behavior when handling the personal care device is natural or non-natural behavior. More particularly, the personal care device or an external device may include a behavior determination algorithm implemented in a control unit of the personal care device in terms of, e.g., software executed by a microprocessor, for automatical determination of natural or non-natural behavior depending on the comparison of the detected real-time data of the user's behavior parameter to the historical data of the user's behavior parameter and/or depending on the detected emotional parameter.
Such non-natural behavior may include an extra-physical effort, e.g. pressing the personal care device's working head harder onto the body surface and/or gripping the handle with increased pressure, and/or unergonomic movements or body positions such as holding the wrist at an extreme angle, and/or additional actions or additional movements such as facial grimaces, skin stretching, and/or an extra-mental effort such as extra concentration with longer periods of not closing the eye's lid and/or using the product for longer in general or in specific areas and/or for specific tasks. In addition or in the alternative, such non-natural behavior may include a change in behavior driven by not getting the end result that a user wants such as missed hairs with the user's natural behavior.
In contrast, natural behavior may be identified as any behavior not including the aforementioned non-natural behaviors, and/or a lower effort than non-natural behaviors and/or intuitive behaviors and/or behaviors more enjoyable than non-natural behaviors, e.g. going along with smiling and/or relaxed gripping of the handle. In addition or in the alternative, natural behavior may be behavior without change and/or behavior with changes that are driven by physiology, e.g. when a shaver is guided with lighter pressure over the Adam's apple.
Due to such automatic determination of natural or non-natural behavior, significant advantages and consumer benefits may be achieved which may include:
Basically, a plurality of different behavioral parameters may be detected during a treatment session and compared to historical data thereof to reliably determine natural behavior and non-natural behavior. According to an advantageous aspect, such behavioral parameters detected in real-time during a treatment session may include skin contact pressure with which the working head is pressed against a body surface as detected by a skin contact pressure detector, and/or stroke frequency and/or stroke direction and/or stroke length at which the personal care device is guided along a body surface as detected by, e.g., an accelerometer, and/or twisting of the personal care device as detected by, e.g., a gyrometer, and/or hair removal activity.
All such real-time data may be compared to historical data of corresponding behavioral parameters, wherein such historical data may come from different sources, e.g. from a storage storing the data of the behavioral parameter as detected by the detector during previous treatment sessions and/or a previous part of a current session, and/or behavioral data detected for the same user when handling other devices, and/or average behavioral parameters from other users for the same type of personal care device and/or other personal care devices, and/or behavioral parameters stored in a database and known to typically indicate non-natural behavior and/or behavioral parameters stored in a database and known to typically indicate natural behavior.
So as to make the identification of non-natural behavior from said comparison of real-time data to historical data easier, the historical data stored in a storage or database may be classified to represent typical natural behavior and/or typical non-natural behavior, wherein such stored historical data may be associated with corresponding classifiers. For example, the behavior determination algorithm may analyze the real-time data of the behavioral parameter in terms of, e.g., value, changes in the signal, increasing and/or decreasing tendency, maximum and minimum values, amplitude, mean value or signal pattern and/or data pattern such as data values over time, and may compare such real-time data and analysis thereof to stored historical data to identify the closest historical set of data to determine natural or non-natural behavior on the basis of the classifier associated with the closest set of data.
In addition or in the alternative, other comparisons such as determination of a deviation of the real-time data from average historical data may be used to identify natural/non-natural behavior.
The emotional data detected by the at least one emotion sensor may be analyzed in a similar manner Preferably real time emotional data is used. Alternatively or in addition, historical data of at least one emotional parameter may be stored and possibly classified as being indicative of natural behavior and/or non-natural behavior so that the behavior determination algorithm may compare the real-time emotion data to such historical emotion data and determine natural or non-natural behavior from the classifier of the closest historical set of data and/or from deviations from stored historical emotion data. Other types of comparisons and/or analyses of the emotion data are possible.
In addition to the aforementioned behavioral parameters and the at least one emotional parameter, the algorithm may further consider environmental parameters as detected by at least one environmental detector and/or at least one physiological detector such as air humidity, skin humidity and/or beard length, when determining natural behavior and/or non-natural behavior.
These and other advantages become more apparent from the following description giving reference to the drawings and possible examples.
According to an aspect, it is suggested to compare the real-time data of the at least one behavioral parameter as detected by the at least one detector when handling the personal care device during a personal care treatment to historical data of at least one behavioral parameter as stored in a storage and/or to detect at least one emotional parameter by means of an emotion detector indicative of the user's emotion when handling the personal care device during a personal care treatment session, and to determine from said comparison of the real-time data of the user's behavior to the historical data thereof and/or to the detected emotional parameter whether the user's behavior when handling the personal care device is natural or non-natural behavior. More particularly, the personal care device may include a behavior determination algorithm implemented in a control unit of the personal care device or in an external device (e.g. in a cloud) in terms of, e.g., software executed by a microprocessor, for automatic determination of natural or non-natural behavior depending on the comparison of the detected real-time data of the user's behavior parameter to the historical data of the user's behavior parameter and/or depending on the detected emotional parameter.
Such non-natural behavior may include extra-physical effort, e.g. pressing the personal care device's working head harder onto the body surface and/or gripping the handle with increased pressure, and/or unergonomic movements or body positions such as holding the wrist at an extreme angle, and/or additional actions or additional movements such as facial grimaces, skin stretching, and/or an extra-mental effort such as extra concentration with longer periods of not closing the eye's lid and/or using the product for longer in general or in specific areas and/or for specific tasks. In addition or in the alternative, such non-natural behavior may include a change in behavior driven by not getting the end result that a user wants such as missed hairs with the user's natural behavior.
In contrast, natural behavior may be identified as any behavior not including the aforementioned non-natural behaviors, and/or a lower effort than non-natural behaviors and/or intuitive behaviors and/or behaviors more enjoyable than non-natural behaviors, e.g. going along with smiling and/or relaxed gripping of the handle. In addition or in the alternative, natural behavior may be behavior without change and/or behavior with changes that are driven by physiology, e.g. when a shaver is guided with lighter pressure over the Adam's apple.
Basically, a plurality of different behavioral parameters may be detected during a treatment session and compared to historical data thereof to reliably determine natural behavior and non-natural behavior. According to an advantageous aspect, such behavioral parameters detected in real-time during a treatment session may include skin contact pressure with which the working head is pressed against a body surface as detected by a skin contact pressure detector, and/or stroke frequency and/or stroke direction and/or stroke length at which the personal care device is guided along a body surface as detected by, e.g., an accelerometer, and/or twisting of the personal care device as detected by, e.g., a gyrometer, and/or hair removal activity.
As determination of natural or non-natural behavior may be a rather complex issue involving a plurality of facets, it may be helpful to detect one or more other behavioral parameters and/or use one or more other behavioral detectors. In particular, non-natural behavior may involve the following aspects and consequently, the following parameters may be detected by the following detectors:
All such real-time data may be compared to historical data of corresponding behavioral parameters, wherein such historical data may come from different sources, e.g. from a storage storing the data of the behavioral parameter as detected by the detector during previous treatment sessions and/or a previous part of a current session, and/or behavioral data detected for the same user when handling other devices, and/or average behavioral parameters from other users for the same type of personal care device and/or other personal care devices, and/or behavioral parameters stored in a database and known to typically indicate non-natural behavior and/or behavioral parameters stored in a database and known to typically indicate natural behavior.
So as to make the identification of non-natural behavior from said comparison of real-time data to historical data easier, the historical data stored in a storage or database may be classified to represent typical natural behavior and/or typical non-natural behavior, wherein such stored historical data may be associated with corresponding classifiers. For example, when the behavior determination algorithm may analyze the real-time data of the behavioral parameter in terms of, e.g., value, changes in the signal, increasing and/or decreasing tendency, maximum and minimum values, amplitude, mean value or signal pattern and/or data pattern such as data values over time, and may compare such real-time data and analysis thereof to stored historical data to identify the closest historical set of data to determine natural or non-natural behavior on the basis of the classifier associated with the closest set of data.
In addition or in the alternative, other comparisons such as deviation of the real-time data from average historical data may be used to identify natural/non-natural behavior.
In addition or in the alternative, the detected behavioral parameter may be analyzed by the behavior determination algorithm to identify a change of behavior with time. For example, users typically alter their behavior over a period of weeks when they start using a new personal care device such as a shaver. In addition or in the alternative, also a change of behavior during a personal care treatment session may be determined.
The emotional data detected by the at least one emotion sensor may be analyzed in a similar manner For example, historical data of at least one emotional parameter may be stored and possibly classified as being indicative of natural behavior and/or non-natural behavior so that the behavior determination algorithm may compare the real-time emotion data to such historical emotion data and determine natural or non-natural behavior from the classifier of the closest historical set of data and/or from deviations from stored historic emotion data. Other types of comparisons and/or analyses of the emotion data are possible.
More particularly, the behavior determination algorithm may apply emotional response tracking technologies and/or gesture tracking technologies to the detected emotional parameters. For example, such emotional response tracking and/or gesture tracking may detect when the user is satisfied during a treatment session what may be identified as natural behavior, and/or when the user is not satisfied during the treatment session what may be identified as non-natural behavior. In addition or in the alternative, facial tracking technology such as via cameras may be used to identify people's emotional responses via analysis of facial cues such as mini-changes to the angles of the corners of the mouth and/or corners of the eyes.
Furthermore, there may be at least one detector for detecting and/or measuring physiological and/or biometric parameters for example of a face and/or a neck and/or a hand portion, wherein, e.g., pulse, stress hormone level, blood pressure, sweat levels and/or pheromone levels may be detected and analyzed to determine natural behavior and/or non-natural behavior.
In addition to the aforementioned behavioral parameters and the at least one emotional parameter, the algorithm may further consider environmental or situational or physiological parameters as detected by at least one environmental detector such as air humidity, skin humidity and/or beard length, when determining natural behavior and/or non-natural behavior.
When non-natural behavior is identified, basically different steps may be taken. For example, there may be a step of identifying what the user is trying to achieve with his non-natural behavior. This may include detection and/or analysis of situational data, e.g. physiological conditions such as beard length, climatic conditions such as air humidity or comparison data of the user's behavioral patterns to a database, e.g. with labelled behavioral patterns that are known by said database.
In addition or in the alternative, feedback may be given to a user upon determination of non-natural behavior. For example, information may be presented to the user which behavioral parameter has been considered to be indicative of non-natural behavior. For example, feedback information may be given that a user applied too much pressure and/or moved the working head along the body with a velocity too high or too low. In addition or in the alternative, information can be presented what would have been the natural behavior.
Such feedback information may be presented in different ways, wherein e.g. optical and/or acoustic signals may be displayed. For example, colored symbols may be indicated, wherein e.g. green light may indicate natural behavior and red light may indicate non-natural behavior such as too much pressure. The feedback may be e.g. provided verbally (e.g. written, spoken) or visually (e.g, lights/LED's that change color, intensity, frequency, etc., e.g. graph on a smart phone display) or acoustically (e.g. buzzer, beep, spoken instructions) or hepatically (e.g. handle vibration).
In addition or in the alternative to such feedback, the personal care device may be modified upon determination of non-natural behavior, wherein at least one working parameter of the personal care device and/or the setting thereof may be changed.
The working parameters which may be adjusted by an adjustment mechanism, may comprise different physical settings and/or functions of the device affecting the personal care treatment, such as a mechanical setting or mechanical function of the working head and/or of the working tool and/or of a drive unit or drive train of the device. More particularly, a working parameter changing the way the personal care treatment is applied, can be adjusted. Such mechanical settings or functions may include the movability of the working head relative to the handle and/or the operation of one or more working tools such as a long-hair cutter and or short hair cutter or groups thereof and the positions thereof relative to other tools, and/or the temperature of a cooling/heating element for cooling/heating the skin, and/or the operation of a lubricant applicator for applying a lubricant to the body portion to be treated.
For example, the personal care device may have a pivotable suspension of its working head to allow for pivoting of the working head relative to the handle about at least one axis, wherein the adjustment mechanism may be configured to adjust the pivoting stiffness of the working head's suspension and/or the resistance and/or unwillingness of the working head against pivoting movements so as to give the personal care device a more aggressive, performance-oriented handling on the one hand and a more comfortable, smoother handling on the other hand, depending on the user's natural or non-natural behavior. More particularly, the adjustment mechanism may vary the torque and/or force necessary to pivot the working head relative to the handle and/or to achieve a certain pivot angle of the working head deviating from a neutral position thereof when non-natural behavior has been determined.
In addition or in the alternative, the adjustment mechanism may be configured to adjust the angular pivoting range of the working head to allow a larger or smaller maximum angular displacement. The personal care device will give a more aggressive, performance-oriented feeling to the user when the maximum available pivoting angle is smaller, whereas a more comfortable, smoother feeling is provided with a larger maximum pivoting angle.
Such adjustment of the pivoting stiffness and/or the angular pivoting range of the working head may be automatically controlled in response to determination of non-natural behavior and/or in response to at least one behavioral parameter selected from the group of parameters comprising skin contact pressure, velocity at which the personal care device is moved along a body portion to be treated, frequency of strokes, angular orientation of the personal care device relative to the gravitational field and position of fingers gripping the handle and position of the working head relative to the body to be treated. As an example when an over-stretched wrist user position is identified together with a shaver orientation that results in the shaver contacting the face in a way that the head is rotated to a maximum angle in one direction, this non-ergonomic behavior is typically motivated by the need for more control/not wanting the head to swivel away. In this situation, the resistance to the rotation of the head can be significantly increased, so that that higher control is provided without the user needing to stretch their wrist in an uncomfortable way.
In the alternative or in addition, a physiological parameter of the user may be detected by a suitable physiological detector. For example, density and/or length of hairs on a skin portion to be shaved may be detected by a visual or optical sensor such as a camera (which may be provided in the personal care device or external to that, e.g. by a smartphone). Furthermore, skin moisture or skin oiliness may be detected.
In addition to sensor data detected during normal use of the shaver, other pieces of information may be used to determine non-natural or natural behavior. For example, a database of one or more known user adaptions may be used to identify when the particular user is adapting his behavior to the shaver, optionally also including typical adaptions for known physiological and/or climatic conditions, wherein such data base may be based on large-scale consumer research and/or may receive updates during the lifetime of the product. The control unit of the personal care device may compare the individually detected parameters to data from the database to find out if the detected data indicates normal, average behavior and/or normal/average parameters and/or represent an adaptive behavior.
In addition or in the alternative to such reference data from a database or other data/sensed parameters, the determination of natural or non-natural behavior may further be supported by data collected from the user himself/herself. For example, the device or an external device as e.g. a smart phone or other input device may include input means such as a touchscreen to input a user's preferences.
A display device may include at least one display field which is used for displaying information relative to the determination of natural or non-natural behavior. For example, such display field may be configured to display pictograms such as a cascade or row of display points.
In addition or in the alternative to a display provided on the personal care device itself, a display such as a touch display may be provided on a cleaning and/or loading station configured to receive and/or be connected to the personal care device so as to charge the device's battery and/or clean the device, wherein a fluid may be applied to, e.g., the shaver head to clean the shaver. Such cleaning and/or charging station may include a display device configured to communicate with the personal care device at least when the shaver is docked into the station so as to display and/or input the aforementioned information.
As explained above, basically different types of behavioral parameters and/or different emotional parameters and/or different environmental parameters and/or different physiological parameters and/or other parameters may be detected and used in different combinations with each other so as to determine non-natural behavior, wherein also different historical data may be used for such determination. Nevertheless, there are some typical examples of non-natural behavior and corresponding real-time data and historical data indicative of such non-natural behavior. Such typical examples include the following:
a) Consumer research has shown that men typically start each shave with natural behavior. (Examples parameters measured under “behavior” (pressure, strokes) etc. can be found listed above.) This can be a starting point to be able to identify natural behavior for a particular man, however, it is not possible to simply take the starting behavior over a fixed period of time during this shave of a shave to equal the natural behavior for this man for several reasons. Firstly, this natural behavior phase lasts for different lengths of time for different men. Secondly, for men who shave one facial region after the other, this “initial phase” is shorter for each region and occurs new for each new region.
Thus further data input is helpful to be able to identify the “initial” periods of each shave that are the natural behavior.
One possible further input is a position identifier e.g. optical system, accelerometers which enables the device to identify where it is on a man's face and thus whether the man shaves one region after the other of his face/neck or whether he goes over his whole face multiple times during the shave. For this purpose the accelerometer is linked to a microprocessor based analysis and the accelerometer data can e.g. be used to identify long and short strokes. These long and short strokes can then be linked to certain facial areas (e.g. only short strokes are possible under the nose).
Alternatively or in addition, a sensor that indicates the number of hairs cut may also serve this purpose, as when the shaver is moved to a new facial region, the number of hairs cut per time is suddenly higher and then reduces again as the shaving progresses in this area.
Further, databases based on consumer research can be provided that relate certain parameters (e.g. average level of shave pressure for a man) to the typical percentage of time spent in natural and non-natural behavior E.g. a man who shaves with particularly high pressure is likely to spend more time in non-natural behavior.
These two further inputs combined with a noticeable change away from the initial measured behavioral values may be used to identify natural behaviors.
b) A man typically changes his shaving behavior over a few weeks when he starts shaving with a new shaver. However, again identification of this change alone is not sufficient to identify natural and non-natural behavior. Typically one might expect that the behavior during the initial shaves is the natural behavior and the non-natural behavior develops with time as the man alters his behavior to his new shaver. However, it could be that the man had already altered his behavior for his old shaver and continued using this non-natural behavior initially with his new shaver. Over time he then noticed that the non-natural behavior is no longer so necessary with his new shaver and so reverts back to using more natural behavior.
Thus, again at least a second type of input is helpful in order to identify natural and non-natural behaviors.
A database from large scale consumer research with data with parameter values for typical natural and non-natural behaviors may be used. In this way, the behavioral change over the weeks can be compared to this database to identify in which direction the change went.
c) Furthermore, “situational” data can be used to identify what the user is trying to achieve with his non-natural behavior.
When shaving a longer beard e.g. 4 days growth and more, a user will typically adapt his behavior to non-natural behavior to fit the product-physiological (longer beard hairs) situation in that he moves the shaver slower than normal. A typical reason for this is that if the user is not careful, the longer hairs can get caught in the foils and tug, which is painful. This slowing down requires concentration as an extra effort and more time. Automatically raising the trimmers in the shaver head so that the beard hairs now just enter the trimmers and no longer the foils can enable to the user to move the shaver at the normal speed (natural behavior), even with longer beard hairs.
However, as this is a fairly dramatic change to the shaver, it may be advisable to have a second sensor type e.g. optical sensor such as a camera that detects hair length to ensure this is the reason for the change of behavior. Time since last usage is not considered sufficient information as many men use wet razors in addition to electric dry shavers.
d) To identify what the user is trying to achieve with his non-natural behavior a database may be used based on large scale consumer research, optionally with machine learning, that based on multiple sensor inputs shows what the user is typically trying to achieve with certain non-natural behaviors.
E.g. going over a specific facial area many times with higher pressure and shorter strokes (possibly also quicker speed) towards the end of the shave is a typical reaction (non-natural behavior) to missed hairs, although missed hairs can be better shaved with less pressure. The shaver may then adapt itself to reduce the pressure, e.g. via a frame that contacts with the skin and so reduces the pressure on the cutting parts for the same force applied by the user. Alternatively or in addition, feedback may be communicated to the user, either e.g. verbal (E.g.: when trying to shave single remaining hairs, try using less pressure (users typically apply more pressure in such situations, which is counterproductive)) or via a signal such as a LED.
Furthermore, it is known from research and consumer tests, that high air humidity leads to sticky skin which means that the frictional forces between skin and shaving foils/trimmers are increased. This leads to a phenomenon called “stick-slip-effect” where the shaver alternately slips easy over the skin or sticks to the skin. This makes shaving more difficult and uncomfortable. Users react in a variety of ways to this, typically they may adapt their behavior to non-natural behavior to the product-environment situation by reducing the shaving pressure they use. As however a general reduction in shaving pressure can have multiple causes, in this situation an additional air humidity sensor could be used in order that the algorithm can identify the appropriate shaver adjustment for this specific situation, such as increasing the pivoting stiffness of the shaver neck (controlled by e.g. a variable spring pre-load of the swivel head's ability to swivel) to reduce the uncontrolled swiveling of the head caused by the stick-slip.
e) Moreover, a database of one or ideally more known classical non-natural shaving behaviors may be used to identify when the particular user is adapting his behavior to the shaver, optionally also including typical adaptations for known physiological and/or climatic conditions. The database is based on large scale consumer research. The database may optionally receive updates during the lifetime of the product.
Using a database alone has limitations, as the behavior of each individual user can vary significantly from what is “typical”. So, preferably data may also be collected from the user himself and optionally from his environment e.g. via sensor measurement of behavior and optionally additionally sensor measurements of physiological and/or climatic parameters and used to modify the algorithm.
For example, a dry electric shaver cuts the beard hairs best when shaving against the grain. Users typically know this, however they find it difficult to do so in the neck area and in particular flat lying hairs on the neck and make shaving here even more difficult. In response, when shaving the neck area, a man will typically rotate his shaver around it's longitudinal axis (D) and change his grip such that the shavers front side points away from him Additionally, the man then rotates the shaver around an axis (H) that is parallel to the swivel axis. However, it is unergonomic and requires extra effort. The reason he intuitively moves the shaver in this way is that for this situation a light swiveling head is counterproductive. By behaving in this way, the use is able to reduce the swivel movement.
Firstly, the shaver recognizes this typically adapting behavior. This can be achieved by multiple different combinations of different sensors such as an accelerometer and a gyroscope. The use of optical sensors, such as cameras, would be an alternative. This may optionally further be supported by the use of physiological and/or climatic data.
Based on usage and optionally physiological and/or climatic data from a high number of users and optionally the use of machine learning, the algorithm knows which typical data from the accelerometer and gyroscope indicate this behavior. Then, when this behavior is identified, a servo-motor increases the preload of a spring that connects the working head and handle to increase the stiffness of the shaver neck and reduce swiveling of the shaver head.
These and other features become more apparent from the example showing in the drawings. As can be seen from
On one end of the shaver 1, a shaver head 3 is attached to the handle, wherein the shaver head 3 may be slewably supported about one or more slewing axes.
The shaver head 3 includes at least one cutter unit 4 which may include a cutter element or undercutter reciprocating under a shearfoil. The shaver head 3 may also include a long hair cutter 8 as it is shown by
So as to drive such cutter unit 4 and the long hair cutter 8, a drive unit 5 may include a motor that can be received within the handle 2 and can be connected to the cutter unit 4 and the long hair cutter 8 by means of a transmitter or drive train extending from the motor to the cutter unit.
As can be seen from
As can be seen from
As can be seen from
Several working parameters and/or working functions of the shaver 1 can be adjusted by means of an adjustment device 6 which may change mechanical settings and/or operational settings of the shaver such as the pivoting stiffness of the shaver head 3 and the position and/or operation of the long-hair cutter 8. Such adjustment device 6 may include one or more adjustment actuators such as electric motors or electric actors or actors of other types using other forms of energy such as magnetic actors. Such adjustment actuators may be controlled by a control unit 80, wherein such control unit 80 may include an electronic control unit, in particular a micro-controller working on the basis of software stored in a memory.
In addition or in the alternative to controlling such adjustment device 6, the control unit 80 is configured to identify natural and/or non-natural behavior of a user of the shaver 1.
More particularly, said control unit 80 may include a behavior determination algorithm 81 which may be implemented, e.g., in terms of a software component executed by a microcontroller of the control unit 80.
As shown by
The real-time data may be signals of a plurality of detectors which may detect relevant parameters when handling the personal care device during a current personal care treatment session.
Such detectors may include in particular a force detector 41 for detecting the force with which the working head 3 is pressed onto the body surface 30. Such force detector 41 may include various sensing means such as a sensor measuring diving of the working head 3 towards the handle 2, a sensor measuring bending stresses in the handle or a sensor measuring torque and/or load of a motor driving the working tools which are all representative of contact pressure.
In addition to detection of the aforementioned force, or in the alternative to such force detection, various other behavioral and/or environmental and/or physiological parameters may be detected, wherein the aforementioned behavior determination algorithm may determine non-natural behavior using such real-time data of such parameters.
More particularly, the following detectors may be provided:
The shaver 1 further may be provided with a detecting unit for detecting or measuring other parameters relevant to the treatment, wherein such detecting unit may include a voltage and/or current detector for detecting power consumption of the drive unit during shaving and/or a time measurement means for measuring shaving time, for example.
In one embodiment non-natural behavior and natural behavior of the shaver usage is identified by detecting the force applied by the user with which the shaver is placed onto the skin area to be shaved. In this embodiment a motor current based detection system is described in more detail, which shall be representative for this applied skin contact force. The shaver motor's (i.e. the drive motor's driving the hair cutting unit) current is measured by e.g. shunt resistor. Any other direct or indirect motor current detection is also applicable. This motor current can be split in three components, i.e.
The complete motor current may be detected, and the two of the motor current components can be subtracted in order to get the third of the components. In more detail: The idle current is determined before the shaver is put onto the skin, e.g. about 0.1 seconds after the user has switched on the shaver. This time interval in which the idle current is measured may be a selected time point after switching on of the shaver and activation of the drive motor between start (almost 0 seconds) and about 1 minute and e.g. about 0.1 sec. after switching on. The current due to the cutting activity is taken from the time jitter of the motor signal. A detailed description, how to measure this time jitter of the motor signal is disclosed in US20190015998 which is incorporated by reference. The remaining current component due to force may be derived from a look up table in which each of those current values is sorted to a force. This look up table may be stored within the microprocessor of the shaver. The look up table may be created independent from the user upfront in a calibration process and may be part of every new shaver with this embodiment. Thus, in the above first step the force by which the shaver is applied on the skin area to be shaved is determined. In a second step, the force that is determined e.g. in this way is then compared with the typical/historical force, applied by this user. To obtain this value, the average of the force values until the current moment is taken. Such an averaging is e.g. mentioned in US20190061183 which disclosure is herewith as well incorporated by reference. The process of averaging may be done via a slow sliding average of the current force, e.g. with a low pass filter. The non-natural behavior is then detected by comparing the present force value as determined by the above first step with the typical/historical value. The database for the averaging in order to arrive at a typical or historical force may be created by one or more former force values. This number may increase with the number of usages of the shaver by the user. As soon, as the present force is higher than the typical/historical force by a predefined amount, a non-natural behavior may be detected. Every value below the predefined amount may be sorted into the natural behavior category. In one embodiment, the difference in the force value just measured compared to the typical or historical can be greater than 3% based on the historical “natural value”, preferably greater than 5% and preferably greater than 10%. In this embodiment the predefined amount is thus 3% or 5% or 10%.
As can be further seen from
As indicated by block S2B, historical data may include behavioral data from a particular user and/or an average user when handling or using other objects such as door handles, keyboards, showers, shaving cream or other personal care devices such as a toothbrush, wherein such behavioral data may be stored together with a comparison to average typical values from other users to indicate the difference of the behavioral data of a particular user from corresponding data of an average user.
As indicated by block S2C, the historical data also may include values or combination of values of one or more behavioral parameters that are known to typically indicate non-natural behavior, wherein such values and/or combination of values may be stored in an external database, in a cloud or other locations, but also in a storage of the personal care device, wherein such storage may be updated on a regular basis, for example.
As indicated by block S2D, the algorithm 81 also may have access to a database, cloud or storage providing for historical data or values or combination of values of one or more behavioral parameters that are known to typically indicate natural behavior.
Furthermore, as indicated by block S2E, the behavior determination algorithm 81 also may take into account real-time data from sensors or detectors that indicate an emotional response of a user and/or historical data of values or combination of values of emotional parameters that are known to typically indicate non-natural behavior and/or natural behavior.
Furthermore, as indicated by block S4, data from other detectors may be taken into account, e.g. environmental data such as air humidity, skin moisture and/or beard length, wherein such environmental data may be used to optionally determine and/or narrow down likely reasons for non-natural behavior, as indicated by block S5. Such optional determination of a likely reason for non-natural behavior may be part of the behavior determination algorithm 81 and applied/executed upon determination of non-natural behavior, as indicated by the flowchart of
Upon determination of natural behavior or non-natural behavior, the personal care device may react in different ways. As indicated by
In addition or in the alternative, feedback may be given to inform the user his/her behavior was non-natural or natural, as mentioned before.
As indicated by block S8, there may be also a user response. Based on the modification of the shaver and/or the feedback given, the user may or may not change behavior. For example, if the pivoting stiffness was increased, the user may find that he no longer needs to hold his hand/arm in an unergonomic way (as he was doing before in order to prevent the shaver head from swiveling) or if he receives the feedback that he is pressing too hard, then he may press less hard. Alternatively, the user may not change behavior and may continue shaving as before.
As the real-time shaving behavior is detected continuously or repeatedly by the aforementioned plurality of detectors, as indicated by block S1, the change or no change in behavior will be sensed and entered into the behavior determination algorithm 81.
On the basis of the detected real-time parameters and the historical data, the device may determine non-natural behavior in different ways, wherein several further examples comprise the following:
A dry electric shaver cuts the beard hairs best when shaving against the grain. Users typically know this, however they find it difficult to do so in the neck area and in particular flat lying hairs on the neck and make shaving here even more difficult. In response, when shaving the neck area, a user will typically rotate his shaver 1 around it's longitudinal axis (D) and change his grip such that the shavers front side points away from him Additionally, the user then rotates the shaver around an axis (H) that is parallel to the swivel axis, as shown by
Firstly, the shaver 1 recognizes this typically adapting behavior. This can be achieved by at least one or multiple different combinations of different sensors. In this embodiment, the use of an accelerometer and a gyroscope may be advantageous. The use of optical sensors, such as cameras, would be an alternative. This may optionally further be supported by the use of physiological and/or climatic data.
Based on usage and optionally physiological and/or climatic data from a high number of users and optionally the use of machine learning, the algorithm knows which typical data from the accelerometer and gyroscope indicate this behavior. Then, when this behavior is identified, a servo-motor increases the preload of the spring (G) that connects head 3 and handle 2 to increase the stiffness of the shaver neck i.e. pivoting stiffness of the head 3 and reduce ease of swiveling of the shaver head 3.
More particularly, the shaver head 3 which is movable relative to the shaver handle 2 with at least one degree of freedom e.g. in terms of rotation of shaver head 3 with respect to a rotation axis (herein called swivel axis (C)) that oriented orthogonally to the shaver handle's longitudinal axis (D)), wherein the shaver handle 2 is equipped with an accelerometer sensor (E) and a gyroscope. The accelerometer (E) is set up in a way to determine the spatial orientation and movement of the shaver 1 in relation to the surrounding gravitational field. The gyroscope is set up to determine twisting of the shaver 1 about its longitudinal axis. The relative movement of shaver head 3 to the handle 2 is controlled by an actuator (F), in this case a servomotor, which is set up to adjust the preload of a spring (G) that connects the shaver handle 2 to the shaver head 3. In addition, a camera system may also be included that identifies the location of flat lying hairs.
The extent to which the users rotate the shaver 1 about both axes and the speed at which they do this varies greatly, not only between different users but as well between different shaves or even during a shave. Therefore, an algorithm may be provided within the control unit (80) that controls the preload adjustment of the spring (G) based on continuous monitoring of the accelerometer data, calculating gliding average and gliding spread values on different timescales (=with variable probing times). In this way, the shaver reacts individually to the users shaving behavior to achieve a smother, more effortless shave.
Also findings such as numerical data from consumer research (e.g. pressing the shaver harder on the face than normal for an individual user suggests that he is adapting his behavior) may be taken into account for adjusting the shaver. For example, the shaver 1 may collect shave data from a particular user, so learns what his typical behavior is (e.g., each man naturally presses the shaver with his own individual pressure against the skin) and can identify when his behavior varies from this.
The shaver head 3 may be mounted so that it can swivel or tilt relative to the handle 2. A flexible shaving head 3 gives freedom how to hold the device, while enabling good adaptation to different face regions. The shaving head 3 can follow the different contours of checks, neck and jawline. This also ensures that for as much of the time as possible the complete cutting element area is in contact with the skin independent of the angle at which the user holds the shaver (within a certain range). This ensures maximum cutting area contact with the face brings the advantages of better efficiency (a quicker shave) and better skin comfort as the pressing force is spread over a larger area leading to lower pressure on the skin.
However, it has been identified that for certain shave behaviors and/or at certain moments in the shave, a low pivoting stiffness can be disadvantageous. Two examples are listed below:
A typical reaction to these situations is that users will adapt how they hold the shaver 1 in their hand. They change the angle of their hand and the shaver 1 so that the shaver handle 2 lies at an extreme angle such that the head 3 cannot swivel any further. However this is unergonomic and extra effort.
The current solution typically offered for these issues is a manual lock for the shaving head which can be activated. The consumer can decide between the flexible and the locked settings, however this can be inconvenient, is an extra step (again more effort) and consumers often try other alternatives (e.g. holding the head with their fingers).
According to another aspect, there may be automatically adapting the force that resists the swivel movement based on behavioral detection (e.g. detects shaving pressure, detects direction and speed of movements, detects angle of shaver handle, detects which cutting elements have contact to the skin). The algorithm that controls the swivel stiffness may modify itself based on the typical behavior of this particular user that it detects over time.
More particularly, the shaver 1 with a swivel head 3 is equipped with pressure sensor 41 and a sensor 43 that detects directions and speed of motion. One or more cutting elements 4 are spring loaded and carry small magnets 103, cf.
The electronic control unit 80 receives the signals of the hall sensors 104 and the accelerometer. A mathematic function translates the signals into pressure and movement data. E.g. the consumer starts to apply higher shaving pressure than typical the cutting elements 4 are moving deeper into the shaving head 3. Or the movements are faster and shorter. The electronic control unit 80 receives these untypical signals from the hall sensors 104 and the accelerometer and translates it to untypical pressure and movement values. These values are compared with a given matrix of values in real time within the control unit 80 and evaluated to generate the assigned signal for the actuator 113. In this example the spring 112 will be pulled to set a specific stiffness of the swing head 3.
Based on previous usage (e.g. other phases in the same shave and/or previous shaves), the algorithm adjusts the e.g. pressure ranges that are considered to be “low”, “medium” or “high. E.g. for a man who typically shaves with a pressure of 1-2 N, the shaver would learn to consider 2N to be a high pressure for this user, whereas for a man who typically shaves with a pressure of 3-5 N, the shaver would learn to consider 2N to be low pressure for this user.
The self-modifying phase of the algorithm starts with the beginning of the first shave: The electronic of the shaver creates medium values. The more shaves are done, the accurate are the stored typical range.
The shaver body may contain a drive motor 5 and a battery 109. The swing head 3 is mounted on an axis 110 which is mounted on a holder 2 of the shaver body. When asymmetric shaving pressure is applied to the shaving system—means more pressure F1 on one of the both foils than F2 on the other—a torque occurs and the shaving head swings around its axis (10) to align on facial contours. The counterforce of the swinging head is minimized to ensure a good adaptation of the shaving system even when low pressure is applied. A pulling spring 112 is mounted between the lower end of the head and the shaver body. The spring sets the force to swing the head. The stronger the spring is set the harder the head can swing. An actuator 113 is attached to the shaver body and holds the end of the spring. It can set the pre-load of the spring 112 by changing the length of the spring. In neutral actuator position the spring has the lowest pre-load and the swing head can swing very easy. At max. actuation the spring is pulled tight and the shaving head needs more shaving pressure to get moved. The consumer feels a more stiff and rigid system. The actuator can set the spring load step-less between min. and max. actuation position.
According to a still further embodiment, the user may be requested to enter data directly e.g. via a smart phone or another device or directly into the shaver in order to provide the algorithm with additional data. This may be a one-time input e.g. after purchase or be requested on a regular basis. This input can then be used to assess, e.g.:
Alternatively or in addition, the user may be requested to provide feedback about his shave over time. In this way, the algorithm can assess which of the modifications it made to the shaver were successful and further optimize how it reacts.
The data from multiple users can then optionally be collected and used to further refine the algorithm.
Optionally, feedback and/or instructions may also be given to the user. E.g.: when trying to shave single remaining hairs, try using less pressure (users typically apply more pressure in such situations, which is counterproductive)
In another specific example, the algorithm defining the adjustment of the shaver, as described in the previous example, may be a self-modifying classifier (e.g. a neural network). In this case, the outputs of the sensors (e.g. shave pressure, stroke frequency, cutting activity), optionally in combination with further parameters like physiological information from sensors/data entry (e.g. hair density) and/or climate data from sensors (e.g. air humidity), are linked to the input nodes of one or more shaving behavior classifiers. In the subsequent (hidden) layers of the classifier, the signals are processed and combined by a number of differentiating nodes. Finally, the classifier decides if the current shaving behavior, optionally combined with further parameters named above in this paragraph, requires increasing or decreasing of the shaver head retention spring preload and thus a firmer or less firm feel of the shaving system on the skin.
To initially define the classifier, it is trained using labelled shave behavior data of a large number of test shaves in advance (factory level). The system then is able to adjust itself more detailed to the user by learning his specific user behavior and optionally further parameters (user-at-home level) and his reactions to the adjustments made by the system and/or by updating the classifier with a further trained version from a web-based source (cloud level). For the latter, data of many different users and shaves is collected to enlarge the training dataset. Training in this context means that the links between differentiation nodes are adjusted, weighted or added/deleted systematically and automatically in order to improve the classifier performance
According to a further aspect, high air humidity leads to sticky skin which means that the frictional forces between skin and shaving foils/trimmers are increased. This leads to a phenomenon called “stick-slip-effect” where the shaver alternately slips easy over the skin or sticks to the skin. This makes shaving more difficult and uncomfortable. Users react in a variety of ways to this, typically they may adapt their behavior to the product-environment situation by reducing the shaving pressure they use. As however a general reduction in shaving pressure can have multiple causes, in this situation an additional air humidity sensor could be used in order that the algorithm can identify the appropriate shaver adjustment for this specific situation, such as increasing the stiffness of the shaver neck (spring pre-load) to reduce the uncontrolled swiveling of the head caused by the stick-slip.
When shaving a longer beard (e.g. 4 days growth and more), a user will typically adapt his behavior to the product-physiological (longer beard hairs) situation in that he moves the shaver slower than normal. A typical reason for this is that if the user is not careful, the longer hairs can get caught in the foils and tug, which is painful. This slowing down requires concentration (extra effort) and more time. Automatically raising the trimmers in the shaver head so that the beard hairs now just enter the trimmers and no longer the foils can enable to the user to move the shaver at the normal speed, even with longer beard hairs. However, as this is a fairly dramatic change to the shaver, it may be advisable to have a second sensor type (e.g. optical sensor such as a camera that detects hair length) to ensure this is the reason for the change of behavior. Time since last usage is not considered sufficient information as many men use wet razors in addition to electric dry shavers.
| Number | Date | Country | Kind |
|---|---|---|---|
| 18164369.3 | Mar 2018 | EP | regional |
