Multifunctional sofa control method and system based on signal data

Abstract

The invention relates to the technical field of intelligent control, in particular to a multifunctional sofa control method and system based on signal data. Analyzing the control instruction to obtain a specific program control instruction type; when a position program control instruction is received, a position control adjustment factor is obtained according to the position control adjustment prediction model; secondly, optimizing a position control regulation factor according to the spatial matching score, and determining a position program control signal; performing program control on the sofa position according to the position program control signal; and when a temperature program control instruction is received, obtaining a temperature control regulation factor according to the temperature control regulation prediction model, determining a temperature program control signal, and performing program control on the sofa temperature according to the temperature program control signal and the comfort score.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a multifunctional sofa control method and system based on signal data. Background Art

[0002] Compared with traditional sofas, multifunctional sofas are widely popular due to their high space utilization, powerful functionality and good comfort. However, the premise for realizing the many advantages of multifunctional sofas is to accurately control the multifunctional sofa. In order to improve the user experience, effective multifunctional sofa control is crucial.

[0003] At present, the existing technology still has shortcomings in data processing for multifunctional sofa control systems; on the one hand, the existing technology cannot meet the user's position adjustment needs using existing control parameters, does not consider the impact of spatial environment and user behavior on program control, and lacks optimization and processing of program control parameters, which to a certain extent restricts the accuracy and flexibility of sofa position adjustment, making it possible for the adjustment result to fail to fully meet the user's actual needs; on the other hand, the existing technology uses control instructions to adjust the temperature in a limited number of parameters, such as ambient temperature and human body temperature, without additionally introducing user preference parameters, which makes the adjustment result fail to meet the user's expected adjustment result, which will reduce the user's experience.

[0004] Therefore, a multifunctional sofa control method and system based on signal data are proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a multifunctional sofa control method and system based on signal data. When the sofa receives a position program control instruction, the position control adjustment factor obtained by the output of the position control adjustment prediction model based on user behavior is used to update the sofa position data, and a space matching score is obtained according to the updated position data and the environmental position data; then, the position program control signal is determined in combination with the position control adjustment factor adjusted according to the judgment result of the space matching score, and the positions of various parts of the sofa are adjusted according to the position program control signal; when the sofa receives a temperature program control instruction, the temperature program control signal is determined in combination with the temperature control adjustment factor obtained by the output of the temperature control adjustment prediction model, and the sofa temperature is adjusted according to the temperature program control signal and the comfort score obtained by evaluating the sofa pressure data, user behavior data and the real-time temperature of the sofa.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A multifunctional sofa control method based on signal data, comprising:

[0008] When receiving the position program control instruction, the user behavior data and the sofa position data are input into the position control adjustment prediction model based on the user behavior to obtain the position control adjustment factor; the sofa position data is updated by using the position control adjustment factor to obtain the updated position data; and the space matching score is obtained according to the updated position data and the environmental position data;

[0009] adjusting the position control adjustment factor according to the judgment result of the spatial matching score;

[0010] Determine a position program control signal in combination with the adjusted position control adjustment factor, and adjust the positions of various parts of the sofa according to the position program control signal;

[0011] When receiving a temperature program control instruction, the sofa temperature data and the ambient temperature data are input into a temperature control adjustment prediction model to obtain a temperature control adjustment factor; a temperature program control signal is determined in combination with the temperature control adjustment factor, and the sofa temperature is adjusted according to the temperature program control signal and a comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time temperature of the sofa.

[0012] Further, the position control adjustment prediction model based on user behavior includes: an input layer, a feature extraction layer, a position control adjustment prediction layer guided by user behavior, and an output layer;

[0013] The input layer is used to convert the sofa position data into position features;

[0014] The feature extraction layer is used to perform deep extraction of the position features by using deep convolution;

[0015] The position control adjustment prediction layer based on user behavior guidance includes a user behavior guidance branch, an LSTM prediction branch and a fusion branch; the user behavior guidance branch is used to obtain the spatial features of user preferences by combining user behavior features and residual blocks; the LSTM prediction branch is used to predict the time-dependent features related to position regulation; the fusion branch is used to fuse the spatial features and the time-dependent features and output them.

[0016] Furthermore, the user behavior data and the sofa position data are input into a position control adjustment prediction model based on user behavior to obtain a position control adjustment factor; the sofa position data is updated using the position control adjustment factor to obtain updated position data; and the specific process of obtaining a spatial matching score according to the updated position data and the environmental position data includes:

[0017] Using the pre-trained user behavior-based position control adjustment prediction model to perform position control adjustment prediction on the user behavior data and the sofa position data to obtain the position control adjustment factor;

[0018] Using the position control adjustment factor to update the sofa position data, to obtain the updated position data;

[0019] According to the updated position data and the environmental position data, an updated position offset and a maximum position offset as well as an updated extension length and a maximum capacity length are obtained;

[0020] The updated position offset, the maximum position offset, the updated stretch length and the maximum capacity length are evaluated to obtain a spatial matching score; wherein the updated position offset includes an updated position horizontal offset, an updated position vertical offset and an updated position angular offset; and the maximum position offset includes a maximum position horizontal offset, a maximum position vertical offset and a maximum position angular offset.

[0021] Furthermore, the specific process of adjusting the position control adjustment factor according to the judgment result of the spatial matching score includes:

[0022] Acquire the position control adjustment factor output by the position control adjustment prediction model;

[0023] Setting a preset spatial matching threshold and comparing it with the spatial matching score;

[0024] If the spatial matching score exceeds the preset spatial matching threshold, the position control adjustment factor is not adjusted; if the spatial matching score does not exceed the preset spatial matching threshold, the spatial matching score is multiplied by the position control adjustment factor to obtain an updated position control adjustment factor.

[0025] Furthermore, the process of determining a temperature program control signal in combination with the temperature control adjustment factor, and adjusting the sofa temperature according to the temperature program control signal and a comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time sofa temperature includes:

[0026] Inputting the sofa temperature data and the ambient temperature data into the temperature control adjustment prediction model to obtain the temperature control adjustment factor;

[0027] Determine the temperature program control signal in combination with the temperature control adjustment factor, send the temperature program control signal to the temperature control component and adjust the sofa temperature according to the temperature program control signal and the comfort score;

[0028] The process of obtaining the comfort score includes:

[0029] Acquire the sofa pressure data, the user behavior data and the real-time sofa temperature; obtain the number of pressure changes within a period of time and the user preferred temperature data according to the sofa pressure data and the user behavior data;

[0030] The number of pressure changes, the user preferred temperature data and the real-time temperature of the sofa are evaluated to obtain the comfort score.

[0031] A multifunctional sofa control system based on signal data, comprising: an acquisition unit and a processing unit; the acquisition unit is divided into a position acquisition unit and a temperature acquisition unit; the processing unit comprises a position processing unit and a temperature processing unit;

[0032] Wherein, the position acquisition unit includes a position acquisition module, a position adjustment prediction module, a space matching scoring module and a position adjustment update module; the position acquisition module is used to receive the position program control instruction, and obtain the user behavior data, the sofa position data and the environmental position data; the position adjustment prediction module inputs the user behavior data and the sofa position data into the position control adjustment prediction model based on the user behavior to obtain the position control adjustment factor; the space matching scoring module is used to obtain the space matching score according to the updated position data and the environmental position data; the position adjustment update module is used to adjust the position control adjustment factor according to the judgment result of the space matching score;

[0033] The position processing unit is used to determine the position program control signal in combination with the adjusted position control adjustment factor, and adjust the positions of various parts of the sofa according to the position program control signal;

[0034] The temperature acquisition unit includes a temperature acquisition module and a temperature adjustment prediction module; the temperature acquisition module is used to receive temperature program control instructions and acquire user behavior data, sofa temperature data and ambient temperature data; the temperature adjustment prediction module inputs the sofa temperature data and the ambient temperature data into a temperature control adjustment prediction model to obtain a temperature control adjustment factor;

[0035] The temperature processing unit is used to determine a temperature program control signal in combination with the temperature control adjustment factor, and adjust the sofa temperature according to the temperature program control signal and a comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time temperature of the sofa.

[0036] Further, the position control adjustment prediction model based on user behavior used by the position adjustment prediction module includes: an input layer, a feature extraction layer, a position control adjustment prediction layer guided by user behavior, and an output layer;

[0037] The input layer is used to convert the sofa position data into position features;

[0038] The feature extraction layer is used to perform deep extraction of the position features by using deep convolution;

[0039] The position control adjustment prediction layer based on user behavior guidance includes a user behavior guidance branch, an LSTM prediction branch and a fusion branch; the user behavior guidance branch is used to obtain the spatial features of user preferences by combining user behavior features and residual blocks; the LSTM prediction branch is used to predict the time-dependent features related to position regulation; the fusion branch is used to fuse the spatial features and the time-dependent features and output them.

[0040] Furthermore, the specific process of the position adjustment prediction module acquiring the position control adjustment factor and the space matching score module acquiring the space matching score includes:

[0041] Using the pre-trained user behavior-based position control adjustment prediction model to perform position control adjustment prediction on the user behavior data and the sofa position data to obtain the position control adjustment factor;

[0042] Using the position control adjustment factor to update the sofa position data, to obtain the updated position data;

[0043] According to the updated position data and the environmental position data, an updated position offset and a maximum position offset as well as an updated extension length and a maximum capacity length are obtained;

[0044] The updated position offset, the maximum position offset, the updated stretch length and the maximum capacity length are evaluated to obtain a spatial matching score; wherein the updated position offset includes an updated position horizontal offset, an updated position vertical offset and an updated position angular offset; and the maximum position offset includes a maximum position horizontal offset, a maximum position vertical offset and a maximum position angular offset.

[0045] Furthermore, the process in which the position adjustment update module is used to adjust the position control adjustment factor according to the judgment result of the space matching score includes:

[0046] Acquire the position control adjustment factor output by the position adjustment prediction module;

[0047] Setting a preset spatial matching threshold and comparing it with the spatial matching score;

[0048] If the spatial matching score exceeds the preset spatial matching threshold, the position control adjustment factor is not adjusted; if the spatial matching score does not exceed the preset spatial matching threshold, the spatial matching score is multiplied by the position control adjustment factor to obtain an updated position control adjustment factor.

[0049] Furthermore, the temperature processing unit is used to determine the temperature program control signal in combination with the temperature control adjustment factor, and the implementation process of adjusting the sofa temperature according to the temperature program control signal and the comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time sofa temperature includes:

[0050] Acquire the temperature control adjustment factor output by the temperature adjustment prediction module;

[0051] Determine the temperature program control signal in combination with the temperature control adjustment factor, send the temperature program control signal to the temperature control component and adjust the sofa temperature according to the temperature program control signal and the comfort score;

[0052] The process of obtaining the comfort score includes:

[0053] Acquire the sofa pressure data, the user behavior data and the real-time sofa temperature; obtain the number of pressure changes within a period of time and the user preferred temperature data according to the sofa pressure data and the user behavior data;

[0054] The number of pressure changes, the user preferred temperature data and the real-time temperature of the sofa are evaluated to obtain the comfort score.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. The present invention proposes a user behavior-based position control adjustment prediction model for predicting position control adjustment factors; the model uses the user behavior characteristics to guide the model to perform targeted position prediction by using the position control adjustment prediction layer based on user behavior guidance. The user behavior characteristics contain the user's preferred position adjustment information, which can guide the model to perform control adjustment prediction in the direction of the user's preferred position, so as to provide the user with the position information needed, thereby improving the user experience.

[0057] 2. The present invention proposes a spatial matching scoring method for adjusting the degree of adjustment of the position control adjustment factor; the method compares the position data updated by the position control adjustment factor with the environmental position data, and conducts a comprehensive evaluation from the aspects of horizontal offset, vertical offset, angular offset and extension limit to obtain a spatial matching score; the position control adjustment factor can be effectively adjusted using this score, thereby ensuring that the adjusted position information is more accurate and effective.

[0058] 3. The present invention proposes a comfort rating method for adjusting the temperature in combination with a temperature program control signal; the method obtains a comfort rating by evaluating the sofa pressure data, the user preferred temperature in the user behavior data, and the real-time temperature of the sofa; wherein the real-time temperature of the sofa is continuously changed by adjusting the sofa temperature using a temperature control adjustment factor; by comprehensively reflecting whether the adjusted temperature meets the user's needs from the user's preferences, pressure changes, and real-time temperature differences, the user experience can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic flow chart of a multifunctional sofa control method based on signal data of the present invention;

[0060] Figure 2 It is a structural schematic diagram of the position control adjustment prediction model based on user behavior of the present invention;

[0061] Figure 3 The present invention is a structural schematic diagram of a multifunctional sofa control system based on signal data. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] With the rise of smart homes and the continuous improvement of living standards, multifunctional sofas have gradually come into the public eye; compared with traditional sofas, multifunctional sofas are widely welcomed due to their high space utilization, powerful functionality and good comfort. However, the premise for realizing the many advantages of multifunctional sofas is to control the multifunctional sofa. In order to further improve the user experience and improve the functionality of multifunctional sofas, it is crucial to propose an effective multifunctional sofa control method.

[0064] At present, the existing technology still has shortcomings in controlling multifunctional sofas. On the one hand, the existing technology does not adjust the multifunctional sofa to the position and angle preferred by the user based on the user behavior data, which will reduce the user experience. On the other hand, the existing technology uses ambient temperature data and body temperature to control the temperature of the multifunctional sofa. The parameters considered are relatively limited, and the sofa temperature cannot be adjusted to the most comfortable state, which will also reduce the user experience.

[0065] To this end, the present invention provides a multifunctional sofa control method and system based on signal data; when the sofa receives a position program control instruction, the sofa position data is updated using a position control adjustment factor obtained by outputting a position control adjustment prediction model based on user behavior to obtain a space matching score; then, a position program control signal is determined in combination with the adjusted position control adjustment factor, and the positions of various parts of the sofa are adjusted according to the position program control signal; when the sofa receives a temperature program control instruction, a temperature program control signal is determined in combination with a temperature control adjustment factor obtained by outputting a temperature control adjustment prediction model, and the sofa temperature is adjusted according to the temperature program control signal and the comfort score obtained by evaluation, thereby effectively improving the user experience.

[0066] The following will be described in detail in combination with Embodiment 1 and Embodiment 2.

[0067] Embodiment 1

[0068] In order to improve the user experience when adjusting the position and temperature of the sofa, a multifunctional sofa manufacturer introduced a multifunctional sofa control method based on signal data proposed in this application. The process diagram of this method can be referred to Figure 1 , the specific process is as follows:

[0069] Obtain the program control instructions of the multifunctional sofa, parse them, and identify the type of program control instructions;

[0070] Furthermore, the program control instruction types include position program control instructions, temperature program control instructions and other program control instructions; this application mainly optimizes position control and temperature control.

[0071] When receiving the position program control instruction, the user behavior data and the sofa position data are input into the position control adjustment prediction model based on the user behavior to obtain the position control adjustment factor;

[0072] Furthermore, the user behavior data includes user preference data and user control adjustment data; the user preference data includes user preferred position adjustment data and temperature adjustment data; the user control adjustment data includes the amplitude and frequency of user adjustment of position and temperature; the sofa position data includes horizontal, vertical and angular position information;

[0073] Furthermore, the user behavior data is obtained by calling from the historical user database, and the sofa location data is collected based on the location sensor;

[0074] Furthermore, the structure of the position control adjustment prediction model based on user behavior is as follows Figure 2 As shown, it includes: an input layer, a feature extraction layer, a position control adjustment prediction layer based on user behavior guidance, and an output layer;

[0075] The input layer is used to convert the sofa position data and the user behavior data into position features and user behavior features;

[0076] The feature extraction layer is used to perform deep extraction of the position feature by using deep convolution to obtain position space feature;

[0077] The position control adjustment prediction layer based on user behavior guidance includes a user behavior guidance branch, an LSTM prediction branch and a feature fusion layer; the user behavior guidance branch is used to obtain the spatial features of user preferences by combining user behavior features and residual blocks; the LSTM prediction branch is used to predict the time-dependent features related to position regulation; the feature fusion layer is used to fuse the spatial features and the time-dependent features and output them;

[0078] Furthermore, the user behavior guidance branch uses four residual blocks based on user behavior feature embedding to extract the spatial features of user preferences; the process of residual blocks based on user behavior feature embedding can be expressed as:

[0079] ;

[0080] ;

[0081] in, is the output feature of the residual block; Represented as feature fusion operation; Represented as a convolution kernel of size 3×3; Represented as Relu activation function; , and They are the input features of the residual block, the intermediate features input before the feature fusion operation, and the user behavior features; Represented as a convolution kernel of size 1×1; Represented as a channel merge operation.

[0082] Furthermore, the LSTM prediction branch uses 4 LSTM layers to extract time-dependent features.

[0083] By using a location control adjustment prediction model based on user behavior to utilize the location adjustment information preferred by the user to guide the model to make control adjustment predictions in the direction of the user's preferred location, the location information required by the user can be provided, thereby improving the user experience.

[0084] The sofa position data is updated using the position control adjustment factor to obtain updated position data; a spatial matching score is obtained based on the updated position data and the environmental position data;

[0085] Furthermore, the position control adjustment factor is used to update the sofa position data to obtain updated position data; and the specific process of obtaining the spatial matching score according to the updated position data and the environmental position data includes:

[0086] Using the pre-trained user behavior-based position control adjustment prediction model to predict the position control adjustment of the user behavior data and the sofa position data, and obtain the position control adjustment factor;

[0087] The sofa position data is updated using the position control adjustment factor to obtain updated position data;

[0088] According to the updated position data, an updated position offset and an updated extension length are obtained; according to the environmental position data, a maximum position offset and a maximum capacity length are obtained;

[0089] Wherein, the environmental position data is collected by an environmental position sensor;

[0090] The updated position offset, the maximum position offset, the updated stretch length and the maximum capacity length are evaluated to obtain a spatial matching score;

[0091] Among them, the updated position offset is calculated based on the difference between the updated position data and the sofa position data, including the updated position horizontal offset, the updated position vertical offset and the updated position angle offset; the maximum position offset is calculated based on the difference between the sofa position data and the position data of the object closest to the sofa in the space, including the maximum position horizontal offset, the maximum position vertical offset and the maximum position angle offset.

[0092] Furthermore, the calculation formula of the spatial matching score is:

[0093] ;

[0094] ;

[0095] in, It is expressed as a space matching score. The larger the score, the more adjustable space the sofa position can have. Expressed as an exponential function; It is expressed as a position offset weight factor, which is set to 0.5; , and They are respectively the updated position horizontal offset, the updated position vertical offset and the updated position angular offset; , and They are the maximum horizontal offset, the maximum vertical offset and the maximum angular offset respectively; It is expressed as the stretch length weight factor, which is set to 0.5; It is expressed as the updated stretch length; Indicates the maximum capacity length.

[0096] The position data after the position control adjustment factor is updated is compared with the environmental position data for reference, and a comprehensive evaluation is performed from the aspects of horizontal offset, vertical offset, angular offset and extension limit to obtain a spatial matching score. This score can be used to effectively adjust the position control adjustment factor according to the size of the adjustable space, thereby ensuring that the adjusted position information is more accurate and effective.

[0097] The position control adjustment factor is adjusted according to the judgment result of the spatial matching score;

[0098] Furthermore, the specific process of adjusting the position control adjustment factor according to the judgment result of the spatial matching score includes:

[0099] Obtaining a position control adjustment factor outputted by a position control adjustment prediction model;

[0100] Set a preset spatial matching threshold and compare it with the spatial matching score;

[0101] If the spatial matching score exceeds the preset spatial matching threshold, the position control adjustment factor is not adjusted; if the spatial matching score does not exceed the preset spatial matching threshold, the ratio of the spatial matching score and the preset spatial matching threshold is multiplied by the position control adjustment factor to obtain an updated position control adjustment factor.

[0102] Furthermore, the preset space matching threshold is set to 0.75. The setting of this threshold is related to the size of the sofa and the sofa environment and can be adjusted according to actual conditions.

[0103] In order to illustrate the position control adjustment factor adjustment strategy proposed in the present application, three sofa position data, user behavior data and environmental position data from different scenarios are randomly selected, which are recorded as sample one, sample two and sample three; the sample data and user behavior data are input into the position control adjustment prediction model based on user behavior for processing to obtain their respective position control adjustment factors; the sofa position data is updated using the position control adjustment factors of each group to obtain the corresponding updated position data; the maximum horizontal position offset, maximum vertical position offset, maximum position angle offset and maximum capacity length of each group are set according to the environment corresponding to the sample data; at the same time, the spatial matching score of each group is calculated in combination with the calculation formula of the spatial matching score; then, the preset spatial matching threshold of each group is set to 0.78, 0.65 and 0.83 respectively in combination with the sofa environment of each group; according to the comparison result of the spatial matching score and the preset spatial matching threshold, the position control adjustment factor adjustment result of each group is obtained, and the position control adjustment factor adjustment test results are shown in Table 1.

[0104] Table 1. Test results of position control adjustment factor adjustment

[0105]

[0106] The space matching score is used to determine whether the adjustable space is limited, and the space matching score is used to adaptively adjust the position control adjustment factor, thereby ensuring that the range of sofa position adjustment is effectively reduced when the adjustable space is insufficient, avoiding the influence of other objects, thereby improving the user experience.

[0107] Determine a position program control signal in combination with the adjusted position control adjustment factor, and adjust the positions of various parts of the sofa according to the position program control signal;

[0108] Furthermore, the number of position control adjustment factors varies according to the number of areas selected by the user to adjust the sofa; the position control adjustment factors are used to determine the position program control signals of different areas, and the corresponding positions of the sofa are adjusted according to the position program control signals of various parts of the sofa.

[0109] When receiving the temperature program control instruction, the sofa temperature data and the ambient temperature data are input into the temperature control adjustment prediction model to obtain the temperature control adjustment factor; the temperature program control signal is determined in combination with the temperature control adjustment factor, and the sofa temperature is adjusted according to the temperature program control signal and the comfort score obtained by evaluating the sofa pressure data, user behavior data and the real-time temperature of the sofa.

[0110] Furthermore, the sofa pressure data and the sofa real-time temperature are collected by pressure sensors and temperature sensors respectively, and the user behavior data is called from the historical user database;

[0111] Furthermore, the process of determining the temperature program control signal in combination with the temperature control adjustment factor, and adjusting the sofa temperature according to the temperature program control signal and the comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time sofa temperature includes:

[0112] Input the sofa temperature data and the ambient temperature data into the temperature control adjustment prediction model to obtain the temperature control adjustment factor;

[0113] Determine a temperature program control signal in combination with the temperature control adjustment factor, send the temperature program control signal to the temperature control component, and adjust the sofa temperature according to the temperature program control signal and the comfort score;

[0114] The process of obtaining the comfort score includes:

[0115] Obtain sofa pressure data, user behavior data and real-time sofa temperature; obtain the number of pressure changes within a period of time and user preferred temperature data based on the sofa pressure data and user behavior data;

[0116] The number of pressure changes, user preferred temperature data and the real-time temperature of the sofa are evaluated to obtain a comfort score;

[0117] The process of adjusting the sofa temperature according to the temperature program control signal and the comfort score is as follows: if the comfort score exceeds the preset comfort threshold, the temperature program control signal is terminated; if the comfort score does not exceed the preset comfort threshold, the temperature continues to be adjusted.

[0118] Furthermore, the difference between the temperature control adjustment prediction model and the position control adjustment prediction model based on user behavior is that the temperature control adjustment prediction layer in the temperature control adjustment prediction model does not incorporate user behavior features, and the other structures are the same;

[0119] Furthermore, the real-time temperature of the sofa is real-time data after the temperature of the sofa is adjusted according to the temperature control adjustment factor;

[0120] Furthermore, the time for recording the number of pressure changes is set to 20 seconds. Of course, the time setting can be flexibly adjusted according to the age and gender of the user and is not unique;

[0121] Furthermore, the calculation formula of the comfort score is:

[0122] ;

[0123] ;

[0124] in, It is expressed as a comfort score. The larger the score, the more suitable the sofa temperature is for the user's needs. Expressed as the inverse tangent function; It is expressed as a pressure weight factor, which is set to 0.45; is the number of pressure changes; It is expressed as a temperature weight factor, which is set to 0.55; It is expressed as the number of pressure-bearing areas; Expressed as Real-time temperature of each stress area; Represented as user preferred temperature;

[0125] Furthermore, the pressure-bearing areas include the sofa's back cushion, seat cushion, armrests, leg rests and other areas, and the number of the pressure-bearing areas can be flexibly set according to the type of sofa; each area is provided with a pressure sensor and a temperature sensor to measure pressure data and temperature data;

[0126] Furthermore, the setting of the preset comfort threshold is closely related to the sofa environment and needs to be adjusted according to actual conditions.

[0127] In order to illustrate the sofa temperature control and adjustment strategy proposed in this application, three sofa temperature data, user behavior data and ambient temperature data from different scenarios are randomly selected, recorded as data one, data two and data three; the sofa temperature data and ambient temperature data are input into the temperature control adjustment prediction model for processing to obtain the temperature control adjustment factor of each group; the temperature program control signal is determined in combination with the temperature control adjustment factor, and the sofa temperature is adjusted according to the temperature program control signal; the number of pressure changes collected in each group is 10, 8 and 9 respectively; the user preference temperature data of each group are 26°C, 23°C and 25°C respectively; in the temperature adjustment process, the comfort score of each group is obtained by combining the calculation formula of the comfort score; at the same time, the preset comfort threshold of each group is set according to the environment of each group; whether to terminate the temperature adjustment is determined based on the comparison between the comfort score of each group and the preset comfort threshold. The temperature adjustment test results are shown in Table 2.

[0128] Table 2. Temperature adjustment test results

[0129]

[0130] In this embodiment, the comfort score is obtained by evaluating the sofa pressure data, the user preferred temperature in the user behavior data, and the real-time temperature of the sofa; in this way, it is possible to comprehensively reflect whether the adjusted temperature meets the user's needs based on the user's preferences, pressure changes, and real-time temperature differences in different stress areas, thereby effectively improving the user experience.

[0131] Embodiment 2

[0132] The present invention proposes a multifunctional sofa control system based on signal data. The structure of the system can refer to Figure 3 , including: an acquisition unit and a processing unit; the acquisition unit is divided into a position acquisition unit and a temperature acquisition unit; the processing unit includes a position processing unit and a temperature processing unit;

[0133] Wherein, the position acquisition unit includes a position acquisition module, a position adjustment prediction module, a spatial matching scoring module and a position adjustment update module;

[0134] The location acquisition module is used to receive location program control instructions and acquire user behavior data, sofa location data and environment location data;

[0135] The position adjustment prediction module inputs the user behavior data and the sofa position data into a position control adjustment prediction model based on user behavior to obtain a position control adjustment factor;

[0136] Further, the position control adjustment prediction model based on user behavior used by the position adjustment prediction module includes: an input layer, a feature extraction layer, a position control adjustment prediction layer guided by user behavior, and an output layer;

[0137] The input layer is used to convert the sofa position data into position features;

[0138] The feature extraction layer is used to perform deep extraction of the position features by using deep convolution;

[0139] The position control adjustment prediction layer based on user behavior guidance includes a user behavior guidance branch, an LSTM prediction branch and a fusion branch; the user behavior guidance branch is used to obtain the spatial features of user preferences by combining user behavior features and residual blocks; the LSTM prediction branch is used to predict the time-dependent features related to position regulation; the fusion branch is used to fuse the spatial features and the time-dependent features and output them.

[0140] In order to verify the effectiveness of the position control regulation prediction model based on user behavior proposed in this application, 800 groups were randomly selected for training and testing three comparison models and model effectiveness tests were performed, which were recorded as Test 1, Test 2 and Test 3; the three comparison models were the position control regulation prediction model based on user behavior proposed in the present invention, the position control regulation prediction model with the residual block retained after removing the user behavior guidance, and the position control regulation prediction model with the branch guided by user behavior removed; they were marked as Model 1, Model 2 and Model 3 respectively; 500 groups were randomly selected from the 800 groups for model training, and 300 groups were used for testing, and the model effectiveness test results were obtained, as shown in Table 3.

[0141] Table 3. Model validity test results

[0142]

[0143] The spatial matching scoring module is used to obtain a spatial matching score according to the updated position data and the environmental position data;

[0144] Furthermore, the specific process of the position adjustment prediction module acquiring the position control adjustment factor and the space matching score module acquiring the space matching score includes:

[0145] Using the pre-trained user behavior-based position control adjustment prediction model to perform position control adjustment prediction on the user behavior data and the sofa position data to obtain the position control adjustment factor;

[0146] Using the position control adjustment factor to update the sofa position data, to obtain the updated position data;

[0147] According to the updated position data and the environmental position data, an updated position offset and a maximum position offset as well as an updated extension length and a maximum capacity length are obtained;

[0148] The updated position offset, the maximum position offset, the updated stretch length and the maximum capacity length are evaluated to obtain a spatial matching score; wherein the updated position offset includes an updated position horizontal offset, an updated position vertical offset and an updated position angular offset; and the maximum position offset includes a maximum position horizontal offset, a maximum position vertical offset and a maximum position angular offset.

[0149] The position adjustment update module is used to adjust the position control adjustment factor according to the judgment result of the spatial matching score;

[0150] Furthermore, the process in which the position adjustment update module is used to adjust the position control adjustment factor according to the judgment result of the space matching score includes:

[0151] Acquire the position control adjustment factor output by the position adjustment prediction module;

[0152] Setting a preset spatial matching threshold and comparing it with the spatial matching score;

[0153] If the spatial matching score exceeds the preset spatial matching threshold, the position control adjustment factor is not adjusted; if the spatial matching score does not exceed the preset spatial matching threshold, the spatial matching score is multiplied by the position control adjustment factor to obtain an updated position control adjustment factor.

[0154] The position processing unit is used to determine the position program control signal in combination with the adjusted position control adjustment factor, and adjust the positions of various parts of the sofa according to the position program control signal;

[0155] The temperature acquisition unit includes a temperature acquisition module and a temperature adjustment prediction module;

[0156] Wherein, the temperature acquisition module is used to receive temperature program control instructions and acquire user behavior data, sofa temperature data and ambient temperature data;

[0157] The temperature adjustment prediction module inputs the sofa temperature data and the ambient temperature data into the temperature control adjustment prediction model to obtain a temperature control adjustment factor;

[0158] The temperature processing unit is used to determine the temperature program control signal in combination with the temperature control adjustment factor, and adjust the sofa temperature according to the temperature program control signal and the comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time temperature of the sofa.

[0159] Furthermore, the temperature processing unit is used to determine the temperature program control signal in combination with the temperature control adjustment factor, and the implementation process of adjusting the sofa temperature according to the temperature program control signal and the comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time sofa temperature includes:

[0160] Acquire the temperature control adjustment factor output by the temperature adjustment prediction module;

[0161] Determine the temperature program control signal in combination with the temperature control adjustment factor, send the temperature program control signal to the temperature control component and adjust the sofa temperature according to the temperature program control signal and the comfort score;

[0162] The process of obtaining the comfort score includes:

[0163] Acquire the sofa pressure data, the user behavior data and the real-time sofa temperature; obtain the number of pressure changes within a period of time and the user preferred temperature data according to the sofa pressure data and the user behavior data;

[0164] The number of pressure changes, the user preferred temperature data and the real-time temperature of the sofa are evaluated to obtain the comfort score.

[0165] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A multifunctional sofa control method based on signal data, characterized in that: include: When receiving the position program control instruction, the user behavior data and the sofa position data are input into the position control adjustment prediction model based on the user behavior to obtain the position control adjustment factor; the sofa position data is updated by using the position control adjustment factor to obtain the updated position data; and the space matching score is obtained according to the updated position data and the environmental position data; adjusting the position control adjustment factor according to the judgment result of the spatial matching score; Determine a position program control signal in combination with the adjusted position control adjustment factor, and adjust the positions of various parts of the sofa according to the position program control signal; When receiving a temperature program control instruction, the sofa temperature data and the ambient temperature data are input into a temperature control adjustment prediction model to obtain a temperature control adjustment factor; a temperature program control signal is determined in combination with the temperature control adjustment factor, and the sofa temperature is adjusted according to the temperature program control signal and a comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time temperature of the sofa.

2. The multifunctional sofa control method based on signal data according to claim 1, characterized in that: The position control adjustment prediction model based on user behavior includes: an input layer, a feature extraction layer, a position control adjustment prediction layer guided by user behavior, and an output layer; The input layer is used to convert the sofa position data into position features; The feature extraction layer is used to perform deep extraction of the position features by using deep convolution; The position control adjustment prediction layer based on user behavior guidance includes a user behavior guidance branch, an LSTM prediction branch and a fusion branch; the user behavior guidance branch is used to obtain the spatial features of user preferences by combining user behavior features and residual blocks; the LSTM prediction branch is used to predict the time-dependent features related to position regulation; the fusion branch is used to fuse the spatial features and the time-dependent features and output them.

3. The multifunctional sofa control method based on signal data according to claim 1, characterized in that: Inputting the user behavior data and the sofa position data into a position control adjustment prediction model based on user behavior to obtain a position control adjustment factor; updating the sofa position data using the position control adjustment factor to obtain updated position data; The specific process of obtaining the spatial matching score according to the updated position data and the environmental position data includes: Using the pre-trained user behavior-based position control adjustment prediction model to perform position control adjustment prediction on the user behavior data and the sofa position data to obtain the position control adjustment factor; Using the position control adjustment factor to update the sofa position data, to obtain the updated position data; According to the updated position data, an updated position offset and an updated extension length are obtained; according to the environmental position data, a maximum position offset and a maximum capacity length are obtained; The updated position offset, the maximum position offset, the updated stretch length and the maximum capacity length are evaluated to obtain a spatial matching score; wherein the updated position offset includes an updated position horizontal offset, an updated position vertical offset and an updated position angular offset; and the maximum position offset includes a maximum position horizontal offset, a maximum position vertical offset and a maximum position angular offset; The calculation formula of the spatial matching score is: ； in, scoring the spatial match; is an exponential function; is the position offset weight factor; , and are respectively the horizontal offset of the update position, the vertical offset of the update position and the angular offset of the update position; , and are respectively the maximum position horizontal offset, the maximum position vertical offset and the maximum position angular offset; is the stretch length weight factor; extending the length for said updating; The maximum capacity length.

4. The multifunctional sofa control method based on signal data according to claim 1, characterized in that: The specific process of adjusting the position control adjustment factor according to the judgment result of the spatial matching score includes: Acquire the position control adjustment factor output by the position control adjustment prediction model; Setting a preset spatial matching threshold and comparing it with the spatial matching score; If the spatial matching score exceeds the preset spatial matching threshold, the position control adjustment factor is not adjusted; if the spatial matching score does not exceed the preset spatial matching threshold, the spatial matching score is multiplied by the position control adjustment factor to obtain an updated position control adjustment factor.

5. The multifunctional sofa control method based on signal data according to claim 1, characterized in that: The process of determining a temperature program control signal in combination with the temperature control adjustment factor, and adjusting the sofa temperature according to the temperature program control signal and a comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time sofa temperature includes: Inputting the sofa temperature data and the ambient temperature data into the temperature control adjustment prediction model to obtain the temperature control adjustment factor; Determine the temperature program control signal in combination with the temperature control adjustment factor, send the temperature program control signal to the temperature control component and adjust the sofa temperature according to the temperature program control signal and the comfort score; The process of obtaining the comfort score includes: Acquire the sofa pressure data, the user behavior data and the real-time sofa temperature; obtain the number of pressure changes within a period of time and the user preferred temperature data according to the sofa pressure data and the user behavior data; The number of pressure changes, the user preferred temperature data and the real-time temperature of the sofa are evaluated to obtain the comfort score; the calculation formula of the comfort score is: ； in, rate the comfort level; is the inverse tangent function; is the pressure weight factor; is the number of pressure changes; is the temperature weight factor; is the number of pressure-bearing areas; For the The real-time temperature of the sofa in each stress-bearing area; The temperature data is preferred for the user.

6. A multifunctional sofa control system based on signal data, characterized in that: include: An acquisition unit and a processing unit; the acquisition unit is divided into a position acquisition unit and a temperature acquisition unit; the processing unit includes a position processing unit and a temperature processing unit; The position acquisition unit includes a position acquisition module, a position adjustment prediction module, a space matching scoring module and a position adjustment update module; the position acquisition module is used to receive the position program control instruction, and obtain the user behavior data, the sofa position data and the environmental position data; the position adjustment prediction module inputs the user behavior data and the sofa position data into the position control adjustment prediction model based on the user behavior to obtain the position control adjustment factor; the space matching scoring module is used to obtain the space matching score according to the updated position data and the environmental position data; The position adjustment update module is used to adjust the position control adjustment factor according to the judgment result of the spatial matching score; The position processing unit is used to determine the position program control signal in combination with the adjusted position control adjustment factor, and adjust the positions of various parts of the sofa according to the position program control signal; The temperature acquisition unit includes a temperature acquisition module and a temperature adjustment prediction module; the temperature acquisition module is used to receive temperature program control instructions and acquire user behavior data, sofa temperature data and ambient temperature data; the temperature adjustment prediction module inputs the sofa temperature data and the ambient temperature data into a temperature control adjustment prediction model to obtain a temperature control adjustment factor; The temperature processing unit is used to determine a temperature program control signal in combination with the temperature control adjustment factor, and adjust the sofa temperature according to the temperature program control signal and a comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time temperature of the sofa.

7. The multifunctional sofa control system based on signal data according to claim 6, characterized in that: The user behavior-based position control adjustment prediction model used by the position adjustment prediction module includes: an input layer, a feature extraction layer, a user behavior-guided position control adjustment prediction layer, and an output layer; The input layer is used to convert the sofa position data into position features; The feature extraction layer is used to perform deep extraction of the position features by using deep convolution; The position control adjustment prediction layer based on user behavior guidance includes a user behavior guidance branch, an LSTM prediction branch and a fusion branch; the user behavior guidance branch is used to obtain the spatial features of user preferences by combining user behavior features and residual blocks; the LSTM prediction branch is used to predict the time-dependent features related to position regulation; the fusion branch is used to fuse the spatial features and the time-dependent features and output them.

8. The multifunctional sofa control system based on signal data according to claim 6, characterized in that: The specific process of the position adjustment prediction module acquiring the position control adjustment factor and the space matching score module acquiring the space matching score includes: Using the pre-trained user behavior-based position control adjustment prediction model to perform position control adjustment prediction on the user behavior data and the sofa position data to obtain the position control adjustment factor; Using the position control adjustment factor to update the sofa position data, to obtain the updated position data; According to the updated position data and the environmental position data, an updated position offset and a maximum position offset as well as an updated extension length and a maximum capacity length are obtained; The updated position offset, the maximum position offset, the updated stretch length and the maximum capacity length are evaluated to obtain a spatial matching score; wherein the updated position offset includes an updated position horizontal offset, an updated position vertical offset and an updated position angular offset; and the maximum position offset includes a maximum position horizontal offset, a maximum position vertical offset and a maximum position angular offset.

9. The multifunctional sofa control system based on signal data according to claim 6, characterized in that: The process of the position adjustment updating module for adjusting the position control adjustment factor according to the judgment result of the spatial matching score includes: Acquire the position control adjustment factor output by the position adjustment prediction module; Setting a preset spatial matching threshold and comparing it with the spatial matching score; If the spatial matching score exceeds the preset spatial matching threshold, the position control adjustment factor is not adjusted; if the spatial matching score does not exceed the preset spatial matching threshold, the spatial matching score is multiplied by the position control adjustment factor to obtain an updated position control adjustment factor.

10. The multifunctional sofa control system based on signal data according to claim 6, characterized in that: The temperature processing unit is used to determine the temperature program control signal in combination with the temperature control adjustment factor, and the implementation process of adjusting the sofa temperature according to the temperature program control signal and the comfort score obtained by evaluating the sofa pressure data, the user behavior data and the real-time sofa temperature includes: Acquire the temperature control adjustment factor output by the temperature adjustment prediction module; Determine the temperature program control signal in combination with the temperature control adjustment factor, send the temperature program control signal to the temperature control component and adjust the sofa temperature according to the temperature program control signal and the comfort score; The process of obtaining the comfort score includes: Acquire the sofa pressure data, the user behavior data and the real-time sofa temperature; obtain the number of pressure changes within a period of time and the user preferred temperature data according to the sofa pressure data and the user behavior data; The number of pressure changes, the user preferred temperature data and the real-time temperature of the sofa are evaluated to obtain the comfort score.

Images