Computer-implemented method for modifying control mode based on user input

Abstract

A computer-implemented method of modifying a control mode based on a user input, the control mode being designed to control a transmittance value of a variable transmittance optical element as a function of an external illumination, the method comprising the steps of: e1) measuring an initial level of the external illumination, e2) setting a transmittance value of the variable transmittance optical element as an initial value in accordance with the initial level of the external illumination according to the control mode, e3) acquiring a user input for adjusting a parameter related to the transmittance value of the variable transmittance optical element, e4) setting a transmittance value of the variable transmittance optical element to a requested value (RV) based on the user input and according to a manual mode, e5) calculating a confidence value based on the requested value and an initial level of the external lighting, the confidence value representing a likelihood that the requested value is a user preference with respect to the initial level of the external lighting, e6) modifying the control mode or maintaining the control mode unchanged based on the user input and the confidence value.

Description

Technical Field

[0001] The present disclosure relates to the technical field of variable transmittance optical elements.

[0002] The present disclosure more particularly relates to a computer-implemented method for modifying a control mode of a transmittance value of a variable transmittance optical element based on user input. Background Art

[0003] Typically, a variable transmittance optical element is one through which a user may have to or need to see. The variable transmittance optical element may specifically include, for example, eyewear equipped with electrochromic lenses, but may also include sunglasses, ski helmets, welding masks, or event windows, doors or walls of a building.

[0004] Variable transmittance optical elements generally have an adaptive transmittance that can be controlled by means of a dedicated control unit. Light transmittance, or more simply transmittance, is generally referred to as tinting and can change from a transparent state to a dark state. The control unit generally has in a memory data or rules that allow the transmittance of the optical lens to be determined at least according to external lighting, which represents the ambient light around the optical element. Based on the measurement of this external lighting, the transmittance of the optical element can be set to an appropriate value. In fact, the greater the external lighting, the lower the transmittance (the optical element is therefore highly tinted). This function is generally referred to as "automatic mode". The automatic mode is usually predetermined during the development of the optical element.

[0005] The variable transmittance optical element may also be controlled by means of an interface for receiving user input, so that the user can adapt the transmittance of the variable transmittance optical element himself when he wants. By means of this interface, the user can request a higher or lower transmittance than that provided by the automatic mode. This functionality is usually referred to as "manual mode".

[0006] In this context, document WO 2017114759 has proposed to use user input to adapt the automatic mode of electrochromic eyewear through training. Therefore, the automatic mode is personalized to better represent the user's preferences.

[0007] However, the user may also use the manual mode for unique events, such as interacting with other people (by increasing transmittance to establish eye contact) or in the case of a headache (by decreasing transmittance). Such user input may make the automatic mode inaccurate. Summary of the invention

[0008] Therefore, one object of the present disclosure is to provide a method of modifying an automatic mode that is less affected by unique events.

[0009] The present disclosure more particularly proposes a computer-implemented method for modifying a control mode of a variable transmittance optical element based on user input, the control mode being designed to control a transmittance value of the variable transmittance optical element depending on external lighting, the method comprising the following steps:

[0010] e1) measure the initial level of exterior lighting,

[0011] e2) setting the transmittance value of the variable transmittance optical element to an initial value according to the initial level of the external lighting according to the control mode,

[0012] e3) obtaining user input for adjusting a parameter related to the transmittance value of the variable transmittance optical element,

[0013] e4) setting the transmittance value of the variable transmittance optical element to a requested value based on the user input and in accordance with a manual mode,

[0014] e5) calculating a confidence value based on the request value and the initial level of the external lighting, the confidence value representing a likelihood that the request value is a user preference regarding the initial level of the external lighting,

[0015] e6) modifying the control mode or maintaining the control mode unchanged based on the user input and the confidence value.

[0016] Thanks to the confidence value, only relevant user inputs can be considered for modifying the control mode (i.e. automatic mode). In other words, the method allows distinguishing user inputs corresponding to unique events, which can be discarded, from user inputs corresponding to stable preferences, which are considered for modifying the control mode, making it more personalized for the user.

[0017] To classify these unique events, the method according to the present disclosure proposes to calculate one or two confidence values ​​based on correlating the input with external lighting and / or analyzing the repetition of the user input. Each confidence value allows the detection of unique events that are typically associated with social interaction or temporary visual fatigue.

[0018] The confidence value may be calculated, for example, based on a correlation between the initial level of external lighting and the requested value according to a predetermined correlation function.

[0019] Correlation is defined here as a relationship between two values, such as a mathematical or statistical relationship. Correlating the requested transmittance with the external lighting allows detecting a relationship between them. When they are correlated, this means that they do not occur together simply by chance.

[0020] For example, when the confidence value is high, the user input is considered because it most likely corresponds to the user's preference regarding the level of external lighting. On the other hand, when the confidence value is low, i.e., when the input is not or only rarely associated with the external lighting, for example because both the requested value of the transmittance and the external lighting are high, the user input will not be considered for modifying the control mode. In fact, when both the transmittance and the external lighting requested by the user are high, the user input most likely corresponds to a social interaction.

[0021] The confidence value may be calculated, for example, based on subsequent user input obtained for adjusting a parameter related to the transmittance value of the variable transmittance optical element after resetting the transmittance value of the variable transmittance optical element to a value determined based on the control mode.

[0022] Evaluating the repetition of the user input by detecting subsequent input from the user after returning to the control mode also allows filtering unique events. In fact, if the user input corresponds to a stable preference, the user will request a transmittance change again. Therefore, a subsequent input is expected. On the contrary, if there is no subsequent input, the input is likely to be a unique event because the user no longer requires a transmittance change.

[0023] Other advantageous and non-limiting features of the method according to the present disclosure are:

[0024] - in step e6), modifying the control mode or maintaining the control mode unchanged according to one of the following rules: the higher the confidence value, the more the user input is considered for modifying the control mode, the control mode is modified when the confidence value is greater than a predetermined confidence value, and the control mode is maintained unchanged when the confidence value is less than a predetermined confidence value;

[0025] - the control model is established on the basis of the training data set, and wherein modifying the control model comprises including the request value and the initial level of the external lighting into the training data set and updating the establishment of the control model;

[0026] - the method further comprises the step of determining a light transmittance range based on an initial level of external lighting, and wherein the predetermined correlation function comprises checking whether the requested value is contained within the light transmittance range, the confidence value being calculated to be higher when the requested value is contained within the light transmittance range than when the requested value is outside the light transmittance range;

[0027] - the light transmittance range is determined based on at least one of: a previous selection by a user, statistical data representing transmittance values ​​selected by a plurality of users in response to an initial level of external lighting;

[0028] - the resetting is performed at the end of a determined time period, the determined time period starting when the transmittance value of the variable transmittance optical element is set to the requested value, the determined time period being based on at least one of: a predetermined time period, a change in external lighting, a position of the variable transmittance optical element, a battery charge of a battery powering the variable transmittance optical element, removal of the variable transmittance optical element from in front of an eye of a user;

[0029] - a subsequent user input is used to set the transmittance value to a subsequent requested value, the method further comprising the step of determining a subsequent light transmittance range based on the requested value, the confidence value being calculated to be higher when the subsequent requested value is included in the subsequent light transmittance range than when the subsequent requested value is outside the subsequent light transmittance range;

[0030] - the subsequent light transmittance range has a minimum value and a maximum value, the minimum value and the maximum value being proportional to the requested value;

[0031] - the method further comprises measuring a subsequent level of external lighting when a subsequent user input is obtained, the method further comprising the step of determining an illumination range based on the initial level of external lighting, the confidence value being calculated to be higher when the subsequent level of external lighting is contained within the illumination range than when the subsequent level of external lighting is outside the illumination range;

[0032] - the lighting range has a minimum level and a maximum level, the minimum level and the maximum level being proportional to the initial level of exterior lighting;

[0033] - the method further comprises the step of triggering a repeat condition when resetting the transmittance value, and wherein the confidence value is calculated to be higher when a subsequent user input is obtained while the repeat condition is still pending than when a subsequent user input is obtained after the repeat condition has expired;

[0034] - the repetition condition is based on at least one of: a time period, a wearing period of the variable transmittance optical element in front of the user, a change in external lighting;

[0035] -Variable transmittance optical elements include electrochromic lenses or liquid crystal lenses.

[0036] The present disclosure also relates to a system, comprising:

[0037] - variable transmittance optical element;

[0038] - a sensor adapted to measure external lighting;

[0039] - an interface adapted to obtain user input,

[0040] - a control unit programmed to implement a control mode designed to control the transmittance value of the variable transmittance optical element as a function of the external illumination;

[0041] The control unit is programmed to implement a method comprising the following steps:

[0042] e1) measure the initial level of exterior lighting,

[0043] e2) setting the transmittance value of the variable transmittance optical element to an initial value according to the initial level of the external lighting according to the control mode,

[0044] e3) obtaining user input for adjusting a parameter related to the transmittance value of the variable transmittance optical element,

[0045] e4) setting the transmittance value of the variable transmittance optical element to a requested value based on the user input and in accordance with a manual mode,

[0046] e5) calculating a confidence value based on the request value and the initial level of the external lighting, the confidence value representing a likelihood that the request value is a user preference regarding the initial level of the external lighting,

[0047] e6) modifying the control mode or maintaining the control mode unchanged based on the user input and the confidence value. DETAILED DESCRIPTION

[0048] The following description is enriched with the accompanying drawings, which are considered as non-limiting examples and will facilitate the understanding of the present disclosure and see how it can be implemented.

[0049] In the attached picture:

[0050] - Figure 1 is a schematic diagram of system 1;

[0051] - Figure 2 is a block diagram presenting the main steps of the method;

[0052] - Figure 3 is a graphical representation of the transmission value of a variable transmittance optical element as a function of external illumination;

[0053] - Figure 4 is to present the first embodiment Figure 2 A block diagram of the steps of the method;

[0054] - Figure 5 are graphical representations of two functions presenting transmission values ​​of a variable transmittance optical element as a function of external illumination;

[0055] - Figure 6 is to present the second embodiment Figure 2A block diagram of the steps of the method.

[0056] Figure 1 Schematic diagram of system 1 is shown. System 1 includes a variable transmittance optical element. Figure 1 As presented, the variable transmittance optical element comprises an eyewear 10 comprising a frame 2 and at least one lens 3, here two lenses 3 mounted in the frame 2. The eyewear 10 is designed to be worn by a user.

[0057] As a variation, the variable transmittance optical element comprises any element through which a user may have to or need to see, such as sunglasses, a ski helmet, a windshield, a window, a door in a wall, a welding helmet.

[0058] Each lens 3 is here an ophthalmic lens. Each lens 3 is designed to be placed in front of one eye of a user of the system 1. Each lens 3 may be corrective or non-corrective. Only one lens 3 will be described below and will be referred to as a "lens".

[0059] Here, the lens 3 has a variable transmittance, which means that the transmittance value of the lens 3 can be controlled. The transmittance value of the lens 3 characterizes the light transmittance of the lens 3, that is to say its ability to transmit light. Classically, light transmittance is the opposite of light absorptivity, and the transmittance value is calculated as the ratio of the light intensity passing through the lens 3 to the light intensity received by the lens 3. The light transmittance can also be referred to as the transmittance level of the lens 3. The transmittance value is here expressed as a percentage.

[0060] In order to have a variable transmittance, the lens 3 is here electrochromic, which means that its transmittance value can be controlled by means of a voltage. The electrochromic lens 3 can be, for example, those described in WO 2018 / 234515, wherein the lens comprises a layer of an electrochromic medium (such as a liquid crystal), a composition comprising an oxidizing compound and a reducing compound in a solvent or in a solid electrochromic layer, and electrodes for applying a voltage to the layer in order to change its transmittance. The lens 3 may also comprise filters and / or polarizers that can be controlled by a voltage. Here, the eyewear 10 comprises a battery, which is for example mounted in the frame 2 and is suitable for providing this voltage.

[0061] As a variant, the lenses are liquid crystal lenses. Liquid crystal technology also allows the transmittance value of the lenses to be controlled by means of an electrical voltage.

[0062] For example, the transmittance value of the electrochromic lens 3 can be set to a predetermined number of values, such as four values ​​ranging from transparent to dark 95%, 50%, 25%, 12%. The transmittance value can also be controlled so that it varies substantially continuously. In any case, the transmittance value is a physical property of the lens 3 and can take different values, such as the "initial", "request" or "subsequent request" values ​​hereinafter.

[0063] like Figure 1 As shown, the system 1 further comprises a sensor 4, a memory 5, an interface 6 and a control unit 7. Here, the sensor 4, the memory 5, the interface 6 and the control unit 7 are located in or on the frame 2, so the system 1 is an independent optical product.

[0064] The sensor 4 is adapted to measure external lighting. Such external lighting represents light originating from outside the system 1, for example outdoor sunlight or indoor artificial light. The sensor 4 is more particularly adapted to provide an electrical signal representative of the intensity of the light it receives. The sensor 4 is, for example, a photodiode, a photoresistor or a phototransistor. The sensor 4 is here mounted in or on the frame 2. When the user wears the eyewear 10, the external lighting now corresponds to the ambient light around the user or the ambient light of the environment located in front of the user, that is to say from in front of the user's face. The sensor can be integrated in the control unit 7. The sensor 4 can also be remote.

[0065] The sensor 4 is more particularly adapted to measure different values ​​of external lighting, hereinafter referred to as levels of external lighting, such as hereinafter "initial" or "subsequent" levels. In other words, the sensor 4 is adapted to quantify the light intensity it receives. The external lighting level is for example expressed in lux.

[0066] The memory 5 (here a computer-readable storage medium) comprises instructions allowing the operation of a control mode, also referred to as automatic mode. The control mode comprises the relationship between the transmittance value of the lens 3 and the external lighting measured by the sensor 4. The control mode allows the transmittance value of the lens 3 to be adapted to the external lighting in an autonomous manner (that is to say without any intervention by the user). To this end, the control mode is designed to determine the transmittance value of the lens 3 as a function of the external lighting. To prevent glare, the higher the external lighting, the lower the transmittance value (when the lens 3 is relatively dark or tinted).

[0067] The control unit 7 is programmed to implement the control mode. The control unit 7 includes at least one processor. The control unit 7 is, for example, a computing module, a computer, a mobile device (such as a smart phone), a connected watch, or a medical device (such as a blood testing device). The control unit 7 may include a virtual machine located on a cloud network or on a server remote from the eyewear 10.

[0068] Here, the control mode is stored in the form of computer-readable data in the memory 5. These computer-readable data represent, for example, mathematical functions, which can be obtained, for example, by Figure 3 The curve C in FIG. 1 is represented. For each level of external illumination in the input, the mathematical function gives a value of the transmittance value in the output. The control pattern can also be stored as a value table or an algorithm, for example in the form of a neural network. Here, the memory 5 is accommodated in the frame 2.

[0069] On the one hand, the control mode is initially predetermined, but on the other hand, the control mode can be modified to better suit the user's preferences. This means that when the user first uses the eyewear 10, the control mode is already stored in the memory 5. For example, the control mode is established when the system 1 is conceived. Classically, the control mode is, for example, initially designed to suit as many potential users as possible. However, in the context of the present disclosure, the control mode can be modified, as described in detail later, in order to make it more personalized or customized to the user.

[0070] The interface 6 is adapted to obtain user input, that is to say input from the user. The interface 6 is thus a man-machine interface. Here, the interface 6 can be manually operated by the user. The interface 6 comprises, for example, a touch-sensitive surface or a mechanical device, such as a button or a rotating edge. Here, the interface 6 is mounted on the frame 2, for example on the temple.

[0071] As a variant, the interface may be a microphone and the user input may be a voice command.

[0072] In any case, the interface 6 is suitable for receiving user inputs, that is to say instructions from a user, regarding the transmittance value of the lens 3. The interface 6 allows implementing a manual mode of the system 1. The manual mode corresponds to an "on demand" operating mode of the user.

[0073] Thanks to the manual mode, the user is able to adapt the transmittance value himself. Therefore, the manual mode comprises a relationship between the user input and the transmittance value in the sense that the manual mode allows the user input to be converted into a transmittance value. By way of example, the interface 6 may comprise a plus button and a minus button, and the manual mode may then operate as follows: when the user presses the plus button or respectively the minus button, the transmittance value is incremented or decremented by a predetermined step. For example, the manual mode determines that pressing the plus button increases the transmittance value of the lens 3 by +10%, and pressing the minus button decreases the transmittance value of the lens 3 by -10%.

[0074] The manual mode is implemented at each user input. The manual mode is typically used when the user wants a transmittance value different (ie higher or lower) than the transmittance value given by the automatic mode. The manual mode is also stored in the memory 5.

[0075] The control unit 7 is particularly adapted to set the transmittance value of the lens 3. On the one hand, the control unit 7 is adapted to set the transmittance value of the lens 3 depending on the external lighting according to a control mode. On the other hand, the control unit 7 is adapted to set the transmittance value of the lens 3 according to a user input. This means that the control unit 7 is also programmed to implement a manual mode when a user input is acquired via the interface 6. In order to set the transmittance value of the lens 3, the control unit 7 is for example adapted to control the voltage supplied to the lens 3.

[0076] like Figure 1 As shown, the control unit 7 is also suitable for controlling the sensor 4 and the interface 6, in particular in the sense that the control unit 7 receives the latest provided signal, and for accessing the memory 5. To this end, the control unit 7 comprises connection means with the other elements of the system 1, in particular with the lens 3, the sensor 4, the memory 5 and the interface 6. The control unit 7 is accommodated in the frame 2 here.

[0077] The system 1 is illustrated here as a stand-alone optical product. However, as a variant, one or more elements of the system can be withdrawn, such as the interface in particular. For example, the interface can be the touch surface of a smartphone. The interface can be a computer graphical user interface equipped with an operating system of the control unit. The interface can be part of a non-control unit on the same housing. Still in the example, the memory and / or the control unit can be partially or completely located on a remote server. At this time, the operations performed by the control unit 7 can be partially or completely performed on the remote server. At this time, the system can include a wireless connection device to establish communication between its different elements.

[0078] The control unit 7 is specifically programmed to implement the Figure 2 The memory 5 comprises instructions which, when executed by the control unit 7, cause the control unit 7 to perform the steps of the method.

[0079] System 1 may also include a detection device (not presented) adapted to determine whether eyewear 10 is worn.

[0080] More specifically, the control unit 7 is configured to implement the following main steps of the method:

[0081] - step e1), measuring the initial level of external lighting,

[0082] - step e2), setting the transmittance value of the electrochromic ophthalmic lens 3 to an initial value according to the control mode,

[0083] - step e3), obtaining a user input for adjusting a parameter related to the transmittance value of the lens 3, which is called the initial user input,

[0084] - step e4), setting the transmittance value of the electrochromic ophthalmic lens 3 to the requested value according to the manual mode,

[0085] - step e5), calculating a confidence value based on the requested value of the transmittance and the initial level of external lighting, the confidence value representing the likelihood that the requested value of the transmittance is the user's preference (i.e. the user's preference) with respect to the initial level of external lighting,

[0086] - Step e6), modifying the control mode or maintaining the control mode unchanged based on the initial user input and the confidence value.

[0087] First, steps e1) and e2) are performed, which correspond to operating the system 1 according to the control mode. These steps are performed when the user wears the eyewear 10, that is to say when the lens 3 is being worn in front of the user's eye.

[0088] The measurement of the initial level of external illumination by the sensor during step e1) is here triggered by the control unit 7. This measurement is performed, for example, a few tens of seconds after the user puts on or turns on the eyewear 10. During step e2), the control unit 7 receives the initial level of external illumination determined by the sensor 4 and then sets the transmittance value of the lens 3 based on the control mode, that is to say as a function of the initial level of external illumination.

[0089] Several iterations of steps e1) and e2) may be performed before obtaining an initial user input. For example, the external illumination is measured at regular time intervals and the transmittance value is set accordingly. The control mode here refers to the default mode when no user input is obtained.

[0090] Secondly, steps e3) and e4) are performed, which corresponds to operating the system 1 according to the manual mode. The manual mode is implemented when a user input is obtained.

[0091] Step e3) is at least partly performed by the user himself. The user performs an initial user input. Specifically, when he estimates that the initial value of the transmittance value does not match his current situation, he performs an initial user input to adjust the parameters related to the transmittance value.

[0092] Here, the parameter related to the transmittance value parameter is the transmittance value itself. In other words, an initial user input is performed to adjust the transmittance value of the lens 3.

[0093] As a variant, the parameter associated with the transmittance value may also be an indication of the activity the user is currently performing or intending to perform (such as driving), or an indication about the user's location (for example indicating whether the wearer is indoors or outdoors), the time of year, or the meteorological conditions at the location, and the automatic mode adapts the transmittance value accordingly.

[0094] Therefore, as described above, the control unit 7 sets the transmittance value of the lens 3 to the requested value based on the initial user input and according to the manual mode. At this time, the requested value represents the transmittance value required by the user.

[0095] Here, the manual mode is implemented during a determined time period. After this time period, the system 1 is operated again according to the control mode. By way of example, this time period is based on: i) a predetermined time period, such as one minute to one hour; ii) a change in external lighting, such as greater than a predetermined level; iii) the position of the lens 3, such as based on a GPS signal provided by the user's smartphone; iv) the charge level of the battery powering the electrochromic lens 3, v) the removal of the eyewear 10 from in front of the user's eyes, such as when the user places the eyewear on their head or in a box.

[0096] Thirdly, steps e5) and e6) are performed. Due to steps e5) and e6), the modification of the control mode is based on the relevance of the initial user input in the context of the initial level of the external lighting, via the confidence value.

[0097] More specifically, the confidence value represents the likelihood (i.e., probability) that the requested value of the transmittance value is the user's preference regarding the initial level of external lighting. In the following, such preferences are characterized as "stable," while, in contrast, unique events are characterized as "transient."

[0098] The stable preference is defined here as the modification of the transmittance value selected by the user in order to adapt to a given level of external lighting. Typically, this modification aims to maximize the user's comfort at a given level of external lighting. In other words, the stable preference represents the ideal transmittance value relative to this given level. Therefore, the stable preference represents the fact that the control mode, in particular the control mode for this given level, should be more customized for the user.

[0099] A unique event is defined here as a modification of the transmittance value that is temporary and specific to the user's personal situation. A unique event is therefore independent of external lighting in the sense that it is not caused by external lighting. A unique event is for example a modification of the transmittance value due to social interaction (e.g. to maintain eye contact) or due to visual fatigue (e.g. due to a headache).

[0100] Here, the confidence value is calculated by the control unit 7, which means that it is calculated by the control unit 7 independently of any intervention of the user (except for the initial user input). The confidence value is usually different from the user's feeling. The confidence value is calculated based on the requested value of the transmittance value and the initial level of the external lighting, in the sense that the confidence value depends on and is associated with a pair of "requested value-initial level".

[0101] Here, the confidence value is defined by a number. The confidence value is more specifically a numerical value. In the following, the confidence value may take a value between 0 and 1, where 0 represents the lowest probability that the user input corresponds to a stable preference and 1 represents the highest probability that the user input corresponds to a stable preference. Then, numerical examples related to the confidence value are given with respect to this specific 0 to 1 range. Of course, other numerical ranges may be used, such as 0% to 100%.

[0102] However, the confidence value may also be defined by tokens, words or symbols, as long as these tokens, words or symbols are arranged in a specific order. For example, the confidence value may be calculated as one of the words in a predetermined list, such as: "very low", "low", "medium", "high" and "very high". In any case, the confidence value should be understood herein as indicating a relative value.

[0103] Then, during step e6), the control mode is modified by taking into account the confidence value, or, on the contrary, maintained unchanged. In a particular manner, a user input that is more likely to be a unique event than a stable preference is not taken into account for modifying the control mode, or is taken into account less than another user input that is more likely to be a stable preference. A user input that is more likely to be a unique event may even be discarded, for example erased from the memory 7. It is therefore noteworthy that the control mode is not modified based on user input that is not related to the external lighting, which prevents degradation of the customization of the control mode.

[0104] In an example, when the confidence value is above a predetermined confidence value, the initial user input is classified as a stable preference, and the initial user input is considered to modify the control mode. Conversely, when the confidence value is below a predetermined confidence value, the initial user input is classified as a transient event, and the control mode is maintained unchanged relative to the initial user input. Thus, the confidence value allows for selection of which user inputs to consider for customizing the control mode and which user inputs to exclude.

[0105] The predetermined confidence value is, for example, 0.5. The predetermined confidence value may be closer to the upper end of the range, for example, 0.8, so as to modify the control mode based on user input that is highly likely to be a stable preference. Thus, the method is more selective about the user input that is considered for modifying the control mode, in the sense that fewer user inputs may be considered.

[0106] In another example, the higher the confidence value, the more the initial user input is considered for modifying the control mode. When considered for modifying the control mode, the initial user input is weighted, for example, with a coefficient proportional to the confidence value. The coefficient that weights the user input can also be associated with the confidence value by a monotonic nonlinear relationship (e.g., an exponential relationship). In this example, for several user inputs, calculating each confidence value allows the user inputs to be sorted from most likely to be stable preferences to least likely, and weighted accordingly to modify the model.

[0107] The modification of the control mode is implemented by means of known methods. Modifying the control mode based on initial user input means that the modification is based on at least one of the initial level of external lighting and the requested value of the transmittance value, that is to say only on the initial level, only on the requested value, or on both the initial level and the requested value. The user's activity can also be used to modify the control mode.

[0108] As an example, when the control mode is established on the basis of a training data set, modifying the control mode includes including the requested value of the transmittance value and / or the initial level of the external lighting into the training data set and updating the establishment of the control mode. Here, the control mode "established on the basis of a training data set" refers to a method known as machine learning, such as a method based on a neural network.

[0109] Still in the example, when the control mode is changed from Figure 3 When represented by the curve C in FIG. 1 , the function defining this curve C can be updated based on the position of the initial user input in the two-dimensional space external illumination EI-transmittance value TR. For example, Figure 3 As shown, in the interval where the transmittance value TR changes from its maximum value (95%, transparent state) to its minimum value (12%, dark state), regression is performed so that the curve C fits the points whose coordinates are the initial level IL and the requested value RV. Figure 3 In FIG. 5 , the original control mode, ie before the modification based on the user input, is represented by the dashed curve referred to as C0.

[0110] As a variant, in particular when the regression curve has a limited number of degrees of freedom, the control mode may be modified taking into account only the initial level or only the requested value, the other being associated with a predetermined value of the control mode (e.g. discrete tinting levels such as 95%, 50%, 25%, 12%). For example, the minimum level of external lighting that will trigger the tinting of the lens may be adjusted taking into account only the initial level, regardless of the requested value.

[0111] Steps e5) and e6) may be performed while the user is wearing the eyewear 10. In this way, the control mode is dynamically customized for the user, since the user input represents a stable preference. As a variant, by saving the data in a memory, steps e5) and / or e6) may be performed later when the user is no longer wearing the eyewear 10. It may be more advantageous, especially in terms of computing power, to modify the control mode at once based on several user inputs, rather than modifying the control mode successively based on each user input.

[0112] Advantageously, the method further comprises a step e7) which comprises acquiring environmental data. The environmental data represent the conditions when the eyewear 10 is used, such as the date or the location when the initial user input is acquired or the activity performed by the user. During step e6), the control mode is then also modified based on the environmental data. In this way, the control mode can be even more customized to the user. Alternatively, due to step e7), it is possible to design several specific control modes based on the environmental data, such as a specific control mode for summer and a specific control mode for winter, or a specific control mode for driving, or a specific control mode for outdoor sports activities or specific locations, such as the seaside or a snow field.

[0113] First embodiment

[0114] exist Figure 4 and Figure 5 In a first embodiment of the method according to the present disclosure, step e5) is based on correlation. More specifically, the calculation of the confidence value is based on the correlation between the initial level of external lighting and the requested value of the transmittance value according to a first correlation function.

[0115] The first correlation function is predetermined in this case, in the sense that, before the method according to the disclosure is carried out, the memory 5 comprises correlation data characterizing the first correlation function.

[0116] Here, the correlation data allows for determining a light transmittance range for a given level of external illumination. Here, the correlation data includes a plurality of light transmittance ranges corresponding to a plurality of external illumination levels. A light transmittance range is defined herein as a plurality of transmittance values ​​comprised between a minimum transmittance value and a maximum transmittance value.

[0117] By way of example, for a given level of external lighting, a light transmittance range is determined based on the most frequent value of transmittance values ​​desired by a plurality of users for this given level of external lighting. In other words, the light transmittance range is based on statistical data representing a plurality of values ​​of transmittance values ​​selected by the plurality of users in response to the given level of external lighting. For example, the light transmittance range corresponds to 90% of the most frequent values ​​of transmittance values ​​selected by the plurality of users for the given level of external lighting, and the light transmittance range is surrounded by 5% of the least frequent values ​​corresponding to both the highest and lowest values ​​of the transmittance values. The plurality of users may be selected based on similarity to the user (age, light sensitivity). Of course, the above numerical percentages may be adapted.

[0118] Still in the example, for a given level of external lighting, the light transmittance range can be determined based on the user's previous selections. For example, the correlation data is collected in advance through a dedicated series of tests performed by users equipped with the eyewear 10.

[0119] The correlation data can be graphically represented in two-dimensional space external illumination EI-transmittance value TR, as Figure 5 The combination or sum of the plurality of light transmittance ranges is defined herein as a correlation area A included between the upper curve A1 and the lower curve A2.

[0120] Following the example of determining a range of light transmittance based on multiple users under a given level of external lighting, the upper curve A1 may correspond to values ​​above 95% of the selected values ​​for transmittance values, and the lower curve A2 may correspond to values ​​below 95% of the selected values ​​for transmittance values.

[0121] like Figure 5 As shown, above a certain level of external lighting, the light transmittance range includes only maximum values ​​of transmittance values, such as 90% (where the upper curve A1 and the lower curve A2 overlap). This can be interpreted, for example, that the plurality of users always choose to fully tint the lens 3 above this level. On the other hand, below a certain level of external lighting, the light transmittance range includes only minimum values ​​of transmittance values, such as 0% (where the upper curve A1 and the lower curve A2 also overlap). This can be interpreted, for example, that the plurality of users never choose to darken the lens 3 below this level.

[0122] like Figure 4 As shown, in this first embodiment, step e5) includes a sub-step e501 of determining an initial light transmittance range based on the initial level of external lighting. Here, determining the initial light transmittance range includes selecting a light transmittance range associated with the initial level of external lighting from the plurality of light transmittance ranges in the memory. This selection is graphically represented in Figure 5The initial light transmittance range RA corresponds to the intersection of the correlation area A and the abscissa axis equal to the initial level IL of the external illumination EI.

[0123] In this first embodiment, step e5) continues with sub-step e502) of calculating the confidence value by implementing a first correlation function. Here, the first correlation function comprises checking whether the requested value belongs to the initial light transmittance range, ie verifying that the requested value is contained in the initial light transmittance range.

[0124] During sub-step e502), a confidence value is calculated to be higher when the requested value is contained within the initial light transmittance range than when the requested value is outside the initial light transmittance range.

[0125] For example, when the request value is included in the initial light transmittance range, the confidence value is calculated as 1, and when the request value is outside the initial light transmittance range, the confidence value is calculated as 0.

[0126] about Figure 5 , a graphical representation of the correlation data shown, a first correlation function comprises checking whether a point with coordinates of the initial level IL and the request value RV is contained in the correlation area A. When the correlation area A is reduced, for example by defining it as 60% of the most frequent values ​​of the transmittance values ​​for each level of external lighting, the correlation criterion is more stringent, which means that the probability that the initial user input is outside the correlation area A is higher.

[0127] As a variation of this first embodiment, the first correlation function may include determining the distance between the request value and the center value of the initial light transmittance range. The shorter this distance, the higher the confidence value. In this case, the confidence value may take a continuous value.

[0128] Still in a variant, several light transmittance ranges can be determined for the initial level. For example, a central light transmittance range and a side light transmittance range surrounding the central light transmittance range. The first correlation function can include checking that the request value belongs to each transmittance range. The confidence value can be calculated as 1, for example, when the request value is within the central light transmittance range, as 0.5, when the request value is within the side light transmittance range but not within the central light transmittance range, and as 0, when the request value is not even contained in the side light transmittance range. More light transmittance ranges can be used to even more discretize the values ​​that the confidence value can take. All transmittance ranges can be determined based on multiple user selections, for example, for 90% more frequent values ​​of transmittance values ​​in the side transmittance value ranges and for 70% more frequent values ​​of transmittance values ​​in the central light transmittance range.

[0129] Second embodiment

[0130] exist Figure 6In the second embodiment presented in , step e5) is based on the repetition of the user input. The calculation of the confidence value is more specifically based on the subsequent user input after obtaining the initial user input after returning to the control mode. Typically, the confidence value when another user input is obtained is higher than the confidence value when no other user input is obtained. In fact, if the initial user input corresponds to a stable preference, the user will request a transmittance change again after returning to the control mode. Therefore, subsequent user input is expected. As described later, the confidence value is even higher when the subsequent user input meets a certain condition.

[0131] like Figure 6 As shown, in this second embodiment, step e5) includes a sub-step 511), which includes returning from the manual mode to the control mode. More specifically, sub-step 511) includes resetting the transmittance value of the lens 3 to a value determined based on the control mode after step e4). In sub-step e511), the transmittance value of the lens 3 is reset to an initial value or another value, which is determined based on another measurement result of the external lighting according to the control mode.

[0132] Resetting the transmittance value (ie returning to the control mode) is performed at the end of the above-determined time period. The determined time period is counted from the time when the transmittance value of the lens 3 is set to the requested value (ie from step e4)).

[0133] In this second embodiment, step e5) then proceeds to a sub-step e512) of acquiring, by means of the interface 6, a subsequent user input for adjusting the transmittance value of the electrochromic lens 3. Here, the term "subsequent" means that the subsequent user input is subsequent to the initial user input, in this sense that the subsequent user input is acquired after the initial user input. In step e512), the subsequent user input is used to set the transmittance value to a subsequent requested value according to the manual mode, as described above. When acquiring the subsequent user input, the sensor 4 also measures the subsequent level of external lighting.

[0134] Preferably, according to the second correlation function, it is determined that the confidence value is higher when the subsequent user input is associated with the first user input than when the subsequent user input is not associated with the initial user input.

[0135] In case of stable preferences, for similar levels of external lighting, users are expected to request similar values ​​of transmittance values. In other words, when a subsequent user input is similar to an initial user input, the initial user input is more likely to represent a stable preference. Advantageously, the second correlation function allows defining whether the initial user input and the subsequent user input are similar with respect to external lighting and / or transmittance values ​​of the lens 3.

[0136] To implement the second relevance function, step e5) comprises a sub-step e513) consisting in determining a neighborhood of the initial user input. The second relevance function then consists in checking whether the subsequent user input belongs to a neighborhood of the initial user input.

[0137] Here, the neighborhood is defined relative to the initial user input in terms of both external lighting and transmittance values. The neighborhood more specifically includes the subsequent light transmittance range and the lighting range.

[0138] The subsequent light transmittance range is determined based on the requested value and includes the requested value. The subsequent light transmittance range is, for example, determined so that it has a minimum value and a maximum value, which are proportional to the requested value. As an example, the minimum value is, for example, half of the requested value, and the maximum value is twice the requested value. Still in the example, the minimum value is included between one-tenth of the requested value and the requested value, and the maximum value is included between the requested value and ten times the requested value.

[0139] The lighting range is determined based on the initial level and includes the initial level. The lighting range is determined, for example, so as to have a minimum level and a maximum level, which are proportional to the initial level. As an example, the minimum level is one-third of the initial level, and the maximum level is three times the initial level. Still in the example, the minimum level is included between one-tenth of the initial level and the initial level, and the maximum level is included between the initial level and ten times the initial level.

[0140] like Figure 6 As shown, once the neighborhood of the initial user input is determined in sub-step e513), a confidence value is calculated in sub-step 515) by implementing a second correlation function.

[0141] Checking that the subsequent user input belongs to the neighborhood of the initial user input (i.e., implementing the second correlation function) includes checking that the subsequent request value belongs to the subsequent light transmittance range or the subsequent level belongs to the illumination range. At this time, the confidence value is calculated to be higher when the subsequent request value is included in the subsequent light transmittance range than when the subsequent request value is outside the subsequent light transmittance range. At this time, the confidence value is also calculated to be higher when the subsequent level is included in the illumination range than when the subsequent level is outside the illumination range.

[0142] Preferably, checking that the subsequent user input belongs to a neighbourhood of the initial user input comprises checking that the subsequent request value belongs to a subsequent light transmittance range and that the subsequent level belongs to a lighting range.

[0143] At this time, the confidence value is calculated to be higher when the subsequent request value is included in the subsequent light transmittance range and when the subsequent level is included in the illumination range than when the subsequent request value is outside the subsequent light transmittance range or the illumination range than when the subsequent request value is outside the illumination range. The confidence value is also calculated to be higher when the subsequent request value is included in the subsequent light transmittance range but the subsequent level is outside the illumination range, or when the subsequent request value is outside the subsequent light transmittance range but the subsequent level is included in the illumination range than when the subsequent request value is outside the subsequent light transmittance range and the subsequent level is outside the illumination range.

[0144] For example, when the subsequent request value is included in the subsequent light transmittance range and when the subsequent level is included in the illumination range, the confidence value is calculated as 1. Then, when the subsequent request value is included in the subsequent light transmittance range but the subsequent level is outside the illumination range, or when the subsequent request value is outside the subsequent light transmittance range but the subsequent level is included in the illumination range, the confidence value is calculated as 0.5. When the subsequent request value is outside the subsequent light transmittance range and the subsequent level is outside the illumination range, the confidence value is calculated as 0.

[0145] As a variant, several illumination ranges may be determined. For example, a central illumination range (e.g., ranging from half to twice the initial level) and a side illumination range surrounding the central illumination range (e.g., ranging from one tenth to ten times the initial level). The second correlation function may include checking that the subsequent level belongs to each illumination range. The confidence value may, for example, be calculated as 1 when the subsequent level is contained in the central illumination range, as 0.5 when the subsequent level is contained in the side illumination range but not in the central illumination range, and as 0 when the subsequent level is not even contained in the side illumination transmittance range.

[0146] As a variant, several subsequent light transmittance ranges may be determined. For example, a central subsequent light transmittance range (e.g., ranging from half to twice the requested value) and a side subsequent light transmittance range surrounding the central subsequent light transmittance range (e.g., ranging from one tenth to ten times the requested value). The second correlation function may include checking that the subsequent request value belongs to each subsequent light transmittance range. The confidence value may, for example, be calculated as 1 when the subsequent request value is within the central subsequent light transmittance range, as 0.5 when the subsequent request value is within the side subsequent light transmittance range but not within the central subsequent light transmittance range, and as 0 when the subsequent request value is not even contained within the side lighting transmittance range.

[0147] As described above, the neighborhood may be determined based on the initial level and the initial value, so that the neighborhood depends dynamically on the initial level and the initial value.As a variant, the neighborhood may comprise a predetermined range of external illumination and transmittance values.

[0148] For example, several lighting ranges may be predetermined, such as 100 to 1000 lux, 1000 to 10 000 lux, and 10 000 to 100 000 lux. Then, the second correlation function includes checking whether the subsequent level of the external lighting is included in the same predetermined lighting range as the initial level of the external lighting. At this time, the confidence value is calculated to be higher when the subsequent level is included in the same predetermined lighting range as the initial level than when the subsequent level and the initial level belong to different predetermined lighting ranges.

[0149] In a similar manner, several subsequent light transmittance ranges may be predetermined, such as 12% to 25%, 25% to 50%, and 50% to 95%. Then, the second correlation function includes checking whether the subsequent request value of the transmittance value is included in the same predetermined subsequent light transmittance range as the request value of the transmittance value. At this time, the confidence value is calculated to be higher when the subsequent request value is included in the same predetermined subsequent light transmittance range as the request value than when the subsequent request value and the request value belong to different predetermined subsequent light transmittance ranges.

[0150] like Figure 6 As shown, in this second embodiment, step 5) also includes a sub-step e514 of triggering a repeat condition. Figure 6 As shown, when returning to the control mode, ie when resetting the transmittance value in step e511), the repetition condition is activated.

[0151] In step e515), the confidence value is calculated to be higher when the subsequent user input is obtained while the repeat condition is still pending than when the subsequent user input is obtained after the repeat condition has failed.

[0152] In practice, the initial user input is more likely to represent a stable preference when the subsequent user input is obtained soon after returning to the control mode. In other words, this means that the initial user input and the subsequent user input are more likely to be uncorrelated when the subsequent user input is obtained later after returning to the control mode. Advantageously, the repetition condition allows the time elapsed between the initial and subsequent user inputs to be taken into account in the calculation of the confidence value.

[0153] The repetition condition is more specifically a repetition or time indicator that allows determining a reference moment. This reference moment is then compared with the moment at which the subsequent user input is obtained. The subsequent user input being obtained "while the repetition condition is still pending" means that the moment at which the subsequent user input is obtained precedes the reference moment in time. The subsequent user input being obtained "after the repetition condition has expired" means that the reference moment precedes the moment at which the subsequent user input is obtained in time.

[0154] For example, in step e515), when subsequent user input is obtained while the repeat condition is still pending, the confidence value is calculated as 1, and when subsequent user input is obtained after the repeat condition has expired, the confidence value is calculated as 0.

[0155] Still in the example, the confidence value can also be calculated based on the time elapsed between the moment when the subsequent user input is obtained and the reference moment. For example, the confidence value can be proportional to the time elapsed between the moment when the subsequent user input is obtained and the reference moment, and when the moment when the subsequent user input is obtained coincides with the reference moment, the confidence value drops to zero.

[0156] Here, the repetition condition is predetermined.

[0157] The repetition condition is based on a time period, for example. This time period is, for example, between 1 minute and 1 hour. In this case, the reference time is calculated as the time when the initial user input is obtained plus the time period.

[0158] The repetition condition is, for example, based on a wearing period during which the lens 3 is worn in front of the eye. This wearing period is, for example, comprised between 1 minute and 20 minutes. It is worth noting that this allows taking into account the effective time during which the eyewear 10 is worn. Thus, the repetition condition is not affected by the period during which the user removes the eyewear 10. To this end, a detection device adapted to determine whether the eyewear 10 is worn is activated. In this case, the reference moment is calculated as the moment at which the initial user input is acquired plus the wearing period.

[0159] The repetition condition may also be based on a change in the external lighting. The change is for example a predetermined absolute change, for example comprised between 1000 lux and 10 000 lux, or a predetermined relative level, for example a doubling of the external lighting. This allows taking into account the user's reaction to the event. If the user does not react to such an event, meaning that a subsequent user input occurs after the change, the confidence value is calculated as low, since it is expected that the user would react. In this case, the reference moment is the moment when the change occurred or the moment when the change occurred plus a short time (for example one to five minutes).

[0160] By way of variation, the confidence value can be calculated directly based on the time interval between obtaining the initial user input and obtaining the subsequent user input, that is, without triggering the repeat condition. The shorter this time interval, the higher the confidence value.

[0161] Third embodiment

[0162] In the third embodiment, the calculation of the confidence value is based on both the correlation (as described in the first embodiment) and the repetition of the user input (as described in the second embodiment). In other words, the third embodiment is a combination of the first embodiment and the second embodiment.

[0163] For example, step e5) may be performed twice, once according to the first embodiment and once according to the second embodiment, calculating the first intermediate confidence value according to the first embodiment and calculating the second intermediate confidence value according to the second embodiment.

[0164] Then, the confidence value is determined as a combination of the first intermediate confidence value and the second intermediate confidence value. For example, the first and second intermediate confidence values ​​may range from 0 to 1, and the confidence value is determined as the product of the first and second intermediate confidence values.

[0165] Step e5) according to both implementations of the first embodiment and the second embodiment can be performed in parallel or sequentially.

[0166] Advantageously, executing them sequentially allows determining the cutoff value of the first calculated intermediate confidence value. When the first calculated intermediate confidence value is lower than the cutoff value, another embodiment is not implemented, which saves time and computing power. In this case, the control mode is maintained unchanged. The cutoff value is, for example, comprised between 0 and 0.5.

[0167] As an example, step e5) according to the first embodiment is first implemented). If the first intermediate confidence value is calculated to be 0, step e5) according to the second embodiment is not implemented, which means, for example, that the neighbor is not even determined and the repetition condition is not even triggered. Alternatively, step e5) according to the second embodiment is first implemented, and when the second intermediate confidence value is below the cutoff value, step e5) according to the first embodiment is not implemented.

Claims

1. A computer-implemented method for modifying a control mode of a variable transmittance optical element based on user input, the control mode being designed to control a transmittance value (TR) of the variable transmittance optical element as a function of external illumination (EI), the method comprising the following steps: e1) measuring an initial level (IL) of said exterior illumination (EI), e2) setting the transmittance value (TR) of the variable transmittance optical element to an initial value according to the initial level (IL) of the external illumination (EI) according to the control mode, e3) obtaining user input for adjusting a parameter related to a transmittance value (TR) of the variable transmittance optical element, e4) setting the transmittance value (TR) of the variable transmittance optical element to a requested value (RV) based on the user input and according to a manual mode, e5) calculating a confidence value based on the request value (RV) and the initial level (IL) of the external lighting (EI), the confidence value representing the likelihood that the request value (RV) is a user preference regarding the initial level (IL) of the external lighting (EI), e6) modifying the control mode or maintaining the control mode unchanged based on the user input and the confidence value.

2. The method according to claim 1, wherein: In step e6), the control mode is modified or maintained unchanged according to one of the following rules: - the higher the confidence value, the more the user input is considered for modifying the control mode; - modifying the control mode when the confidence value is greater than a predetermined confidence value, and maintaining the control mode unchanged when the confidence value is less than the predetermined confidence value.

3. The method according to claim 1 or 2, wherein: The control pattern is established on the basis of a training data set, and wherein modifying the control pattern comprises including the request value (RV) and the initial level (IL) of the exterior lighting (EI) into the training data set and updating the establishment of the control pattern.

4. The method according to any one of claims 1 to 3, wherein: Calculating the confidence value is based on a correlation between the initial level (IL) of the exterior illumination (EI) and the request value (RV) according to a predetermined correlation function.

5. The method according to claim 4, comprising the step of determining a light transmittance range based on an initial level (IL) of the external illumination (EI), and wherein: The predetermined correlation function includes checking whether the request value (RV) is included in the light transmittance range, and the confidence value is calculated to be higher when the request value (RV) is included in the light transmittance range than when the request value (RV) is outside the light transmittance range.

6. The method according to claim 4 or 5, wherein: The light transmittance range is determined based on at least one of the following: - the user's previous selections; - statistical data representing transmittance values ​​(TR) selected by a plurality of users in response to an initial level (IL) of said external illumination (EI).

7. The method according to any one of claims 1 to 6, wherein: Calculating the confidence value is based on obtaining subsequent user input for adjusting a parameter related to the transmittance value (TR) of the variable transmittance optical element after resetting the transmittance value (TR) of the variable transmittance optical element to a value determined based on the control mode.

8. The method according to claim 7, wherein: The resetting is performed at the end of a determined time period, the determined time period starting when the transmittance value (TR) of the variable transmittance optical element is set to a requested value (RV), the determined time period being based on at least one of: a predetermined time period, a change in the external illumination (EI), a position of the variable transmittance optical element, a battery charge of a battery powering the variable transmittance optical element, removal of the variable transmittance optical element from in front of the user's eyes.

9. The method according to claim 7 or 8, wherein: The subsequent user input is used to set the transmittance value (TR) to a subsequent request value, and the method further includes a step of determining a subsequent light transmittance range based on the request value (RV), the confidence value being calculated to be higher when the subsequent request value is contained within the subsequent light transmittance range than when the subsequent request value is outside the subsequent light transmittance range.

10. The method according to claim 9, wherein: The subsequent light transmittance range has a minimum value and a maximum value, the minimum value and the maximum value being proportional to the request value (RV).

11. A method according to any one of claims 7 to 10, comprising measuring a subsequent level of the external illumination (EI) when the subsequent user input is obtained, the method further comprising the step of determining an illumination range based on the initial level (IL) of the external illumination (EI), the confidence value being calculated to be higher when the subsequent level of the external illumination (EI) is contained within the illumination range than when the subsequent level of the external illumination (EI) is outside the illumination range.

12. The method according to claim 11, the illumination range having a minimum level and a maximum level, the minimum level and the maximum level being proportional to an initial level (IL) of the exterior illumination (EI).

13. The method according to any one of claims 7 to 12, further comprising the step of triggering a repeat condition when resetting the transmittance value (TR), and wherein, The confidence value is calculated to be higher when the subsequent user input is obtained while the repeat condition is still pending than when the subsequent user input is obtained after the repeat condition has expired.

14. The method according to claim 13, wherein: The repetition condition is based on at least one of the following: - time period, - a wearing period during which the variable transmittance optical element is located in front of the user, - Changes in the external lighting (EI).

15. A system (1), comprising: - variable transmittance optical element; - a sensor (4) adapted to measure external illumination (EI); - an interface (6) adapted to obtain user input, - a control unit (7) programmed to implement a control mode designed to control the transmittance value (TR) of the variable transmittance optical element as a function of the external illumination (EI); The control unit (7) is programmed to implement a method comprising the following steps: e1) measuring an initial level (IL) of said exterior illumination (EI), e2) setting the transmittance value (TR) of the variable transmittance optical element to an initial value according to the initial level (IL) of the external illumination (EI) according to the control mode, e3) obtaining user input for adjusting a parameter related to a transmittance value (TR) of the variable transmittance optical element, e4) setting the transmittance value (TR) of the variable transmittance optical element to a requested value (RV) based on the user input and according to a manual mode, e5) calculating a confidence value based on the request value (RV) and the initial level (IL) of the external lighting (EI), the confidence value representing the likelihood that the request value (RV) is a user preference regarding the initial level (IL) of the external lighting (EI), e6) modifying the control mode or maintaining the control mode unchanged based on the user input and the confidence value.

Images