Intelligent sofa control system based on network technology

Abstract

The invention relates to the field of intelligent home control, and particularly discloses an intelligent sofa control system based on a network technology, which is characterized in that by analyzing pressure distribution, weight distribution, contact area distribution and back inclination angle when a user sits on a sofa in various sitting postures, recognition models of various sitting postures are trained and optimized; the sitting posture recognition model is analyzed from multi-aspect sitting posture characteristics, so that the precision of the sitting posture recognition model is improved; by recognizing the sitting posture of the user, judging whether the sitting posture of the user is poor and whether the user sits for a long time or not and reminding, the user can keep a good sitting posture habit and prevent physical diseases caused by long-time static sitting, and the comfort level and the health level of the user are improved; by acquiring the target posture parameters of the sofa and feeding the target posture parameters back to the control end of the sofa, the posture of the sofa is regulated and controlled, highly customized and personalized services of the intelligent sofa are facilitated, and the best use experience and comfort are provided for a user.

Description

Technical Field

[0001] The invention relates to the field of intelligent home control, and in particular to an intelligent sofa control system based on network technology. Background Art

[0002] As smart homes gradually become an important part of family life, smart sofas, as key products, meet the diverse needs of users in different scenarios by combining technologies such as artificial intelligence and the Internet of Things, providing users with a more comfortable, convenient and healthy life experience. The control of smart sofas not only improves the convenience and comfort of life, but also brings users comprehensive health management and quality of life improvement, which is of great necessity in modern home life.

[0003] However, the existing intelligent sofa control system still has some limitations and shortcomings in practical applications.

[0004] For example, the existing Chinese patent with publication number CN113433834A discloses a smart sofa and smart home control system, including a temperature adjustment control, a massage part and a lifting column. The upper surface of the main body is provided with a temperature adjustment control, the side of the main body is provided with a massage part, and the lower surface of the main body is fixedly connected with a lifting column. The smart sofa is provided with a temperature adjustment control, a massage part and a smart interactive part in the sofa, so that the sofa has the functions of intelligently adjusting the temperature, massaging the back as needed, and connecting with other smart homes. The mounting plate is connected to the control computer, and the sofa is directly controlled from the interactive armrest to achieve temperature adjustment, human body massage and other entertainment or office functions. The backrest rotates a certain angle relative to the main body, so that the device can adapt to different usage modes such as sitting, standing, sitting and lying flat, which expands the use function of the sofa and can also solve the emergency situation of lack of furniture in the family.

[0005] For example, the existing Chinese patent with publication number CN107908124A discloses a control method, control system and smart sofa of a smart sofa, including: when the smart sofa receives a tiling control signal, it detects whether there are obstacles around the smart sofa; when there are no obstacles around the smart sofa, the smart sofa executes the tiling control signal to control the backrest of the smart sofa to be tiled; when there are obstacles around the smart sofa, the smart sofa detects whether the distance between the smart sofa and the obstacle is greater than a preset distance; when the distance between the smart sofa and the obstacle is greater than the preset distance, the smart sofa executes the tiling control signal to control the backrest of the smart sofa to be tiled. The invention detects whether there is enough space for the smart sofa. When there is not enough space around the smart sofa, the smart sofa does not perform the tiling action, thereby reducing the possibility of the smart sofa colliding with obstacles and causing damage to the smart sofa, which is conducive to the long-term use of the smart sofa.

[0006] For example, an existing Chinese patent with publication number CN116300630A discloses a control system for a smart home sofa, including a central processing unit, a temperature adjustment unit, a body sensing unit, an information transmission unit, a massage unit, an adjustment unit and an adjustment unit installed on the smart sofa, wherein the temperature adjustment unit includes a heating module and a cooling module, the body sensing unit includes a temperature sensor for obtaining normal data reference of the body, and the information transmission unit includes a wireless signal transceiver module for information transmission of signal reception and transmission. The body data of the sofa person is sensed by the temperature sensor, so that the temperature of the sofa is regulated in cooperation with the indoor temperature sensor. At the same time, the cooling module is triggered to perform cooling processing when the indoor temperature is high, so that the heart rate detector can detect the heart rate of the user on the smart sofa, thereby adjusting the sofa, which can have a soothing effect on the person, and the heart rate information is analyzed by the data analysis module for storage.

[0007] The above patent has some shortcomings: First, it does not involve the recognition of the user's sitting posture and the reminder of the user to adjust the sitting posture, and therefore does not have the function of encouraging users to maintain good sitting habits and prevent physical diseases caused by long-term sitting, and thus cannot better improve the user's comfort and health level.

[0008] Secondly, it does not have the function of automatic control of the sofa posture, such as automatically adjusting the sofa backrest inclination angle, seat cushion height, seat cushion inclination angle and other sofa parameters according to the user's preferences or needs, which is not conducive to the smart sofa to achieve highly customized and personalized services, and cannot provide users with the best user experience and comfort. Summary of the invention

[0009] In view of the above problems, the present invention proposes an intelligent sofa control system based on network technology to realize the function of intelligent home control.

[0010] The technical solution adopted by the present invention to solve its technical problem is: the present invention provides an intelligent sofa control system based on network technology, including: a sitting posture recognition model training optimization module: used to obtain sitting posture feature information of human bodies of various body types sitting on the sofa in various sitting postures through virtual simulation technology, wherein the sitting posture feature information includes pressure distribution, weight distribution, contact area distribution, and back tilt angle, train recognition models of various sitting postures, further judge whether their accuracy meets the requirements, and optimize the recognition models of various sitting postures whose accuracy does not meet the requirements.

[0011] User sitting posture matching and recognition module: used to collect the sitting posture feature information of the user when sitting on the sofa, combine the recognition models of various sitting postures, analyze the matching degree between the user's sitting posture feature information and various sitting postures, and further identify the user's sitting posture.

[0012] The user's bad sitting posture reminder module is used to compare the user's sitting posture with a set of bad sitting postures, determine whether the user's sitting posture is bad, and give a reminder.

[0013] User sedentary reminder module: used to obtain the length of time the user maintains the same sitting posture and the cumulative length of time the user uses the sofa at a single time, determine whether the user has been sedentary for a long time, and give a reminder.

[0014] Sofa posture automatic control module: used to obtain the startup requirements and working mode of the sofa, where the working mode includes default mode, custom mode, and normal mode, further analyze the target posture parameters of the sofa, and feed them back to the control end of the sofa, thereby controlling the posture of the sofa.

[0015] Database: used to store bad sitting posture sets, sitting time thresholds, and sofa posture parameters in default mode.

[0016] Compared with the prior art, the intelligent sofa control system based on network technology described in the present invention has the following beneficial effects: 1. The present invention obtains the pressure distribution, weight distribution, contact area distribution, and back tilt angle of people of various body shapes sitting on the sofa in various sitting postures, trains recognition models for various sitting postures, determines whether their accuracy meets the requirements and performs optimization, analyzes the sitting posture recognition model from various aspects of sitting posture characteristics, and then improves the accuracy of the sitting posture recognition model, thereby improving the accuracy of the user sitting posture recognition results based on the sitting posture recognition model.

[0017] 2. The present invention collects the sitting posture feature information of the user when sitting on the sofa, combines it with various sitting posture recognition models, identifies the user's sitting posture, further determines whether the user's sitting posture is poor and whether the user has been sitting for a long time, and gives a reminder, which helps the user maintain good sitting habits, prevent physical diseases caused by long-term sitting, and thus improve the user's comfort and health level.

[0018] 3. The present invention obtains the startup requirements and working mode of the sofa, further analyzes the target posture parameters of the sofa, and feeds them back to the control end of the sofa, thereby regulating the posture of the sofa, which is conducive to the smart sofa to achieve highly customized and personalized services, thereby providing users with the best use experience and comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1This is a system module connection diagram of the present invention.

[0021] Figure 2 It is a flow chart of the training optimization module of the sitting posture recognition model of the present invention.

[0022] Figure 3 It is a schematic diagram of the process of the sofa posture automatic control module of the present invention. DETAILED DESCRIPTION

[0023] 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.

[0024] See also Figure 1 As shown, the present invention provides an intelligent sofa control system based on network technology, including a sitting posture recognition model training optimization module, a user sitting posture matching recognition module, a user sitting posture poor reminder module, a user sitting for a long time reminder module, a sofa posture automatic control module, and a database.

[0025] The user sitting posture matching and recognition module is respectively connected to the sitting posture recognition model training and optimization module and the user sitting posture bad reminder module; the user sitting for a long time reminder module is respectively connected to the user sitting posture bad reminder module and the sofa posture automatic control module; the database is respectively connected to the user sitting posture bad reminder module, the user sitting for a long time reminder module and the sofa posture automatic control module.

[0026] The sitting posture recognition model training and optimization module is used to obtain the sitting posture characteristic information of people of various body types sitting on the sofa in various sitting postures through virtual simulation technology, wherein the sitting posture characteristic information includes pressure distribution, weight distribution, contact area distribution, and back tilt angle, train recognition models for various sitting postures, further determine whether their accuracy meets the requirements, and optimize the recognition models for various sitting postures whose accuracy does not meet the requirements.

[0027] Further, see Figure 2 As shown, the specific working process of the sitting posture recognition model training optimization module includes: taking height and weight as variables of human body shape, setting and adjusting the variables of human body shape multiple times according to preset principles, and obtaining the body shapes of various human bodies.

[0028] Through virtual simulation technology, various sitting postures of human bodies of various sizes are simulated on the sofa, and sitting posture characteristic information of human bodies of various sizes sitting on the sofa in various sitting postures is obtained.

[0029] As a preferred solution, the various sitting postures include but are not limited to: sitting upright, sitting with a hunched back, leaning forward, leaning backward, leaning to the left, leaning to the right, crossing the left leg, crossing the right leg, etc.

[0030] As a preferred solution, pressure distribution refers to the relative proportion of pressure in each sub-area of ​​each component of the sofa, wherein the components of the sofa include a headrest, a backrest, a lumbar support, a seat cushion, a leg pad, a footrest, and an armrest. The relative proportion of pressure in each sub-area of ​​each component refers to the ratio between the pressure detected in each sub-area of ​​each component and the cumulative pressure detected in the sub-areas of its component.

[0031] As a preferred solution, the cumulative pressure detected by the component sub-region refers to the sum of the pressures detected by each sub-region in the component.

[0032] As a preferred solution, weight distribution refers to the relative proportion of the weight of each sub-area in each component of the sofa, wherein the relative proportion of the weight of each sub-area in each component refers to the ratio between the detected weight of each sub-area in each component and the cumulative weight detected in the sub-areas of its component.

[0033] As a preferred solution, the contact area distribution refers to the relative proportion of the contact area of ​​each sub-area in each component of the sofa, wherein the relative proportion of the contact area of ​​each sub-area in each component refers to the ratio between the detected human body contact area of ​​each sub-area in each component and the cumulative human body contact area detected by the sub-areas of its component.

[0034] As a preferred solution, the back tilt angle is the angle between the tangent line of the longitudinal contour line of the human back and the horizontal reference line, or the angle between the tangent plane of the human back curved surface and the horizontal reference plane. The range of the back tilt angle is [0,π].

[0035] According to the sitting posture feature information of people of various body types when sitting on the sofa in various sitting postures, the sitting posture feature information of people of various body types when sitting on the sofa corresponding to the various sitting postures is classified and counted, and data sets of various sitting posture recognition models are constructed. The data sets of various sitting posture recognition models are divided according to the preset ratio between the training set and the test set to obtain the training set and test set of various sitting posture recognition models.

[0036] As a preferred solution, the training set and the test set are accumulated to form a data set, and the training set is larger than the test set. In a specific embodiment, the ratio between the training set and the test set is 3:1.

[0037] Based on the training sets of various sitting posture recognition models, machine learning algorithms are used to analyze the pressure distribution, weight distribution, contact area distribution, and back tilt angle corresponding to various sitting postures, and to train recognition models for various sitting postures.

[0038] As a preferred solution, the machine learning algorithm includes but is not limited to: support vector machine, random forest, decision tree, etc.

[0039] As a preferred solution, the training and optimization of the sitting posture recognition model belongs to the production test stage of the sofa, at which time the sofa is not put into use, and the data used to train the sitting posture recognition model is the initial input data of the sofa control system.

[0040] In another specific embodiment, the sitting posture feature information of subjects of various body shapes sitting on the sofa in various sitting postures during the sofa production test phase is obtained to obtain data sets of various sitting posture recognition models, and then the recognition models of various sitting postures are trained.

[0041] Furthermore, the specific working process of the sitting posture recognition model training optimization module also includes: according to the test sets of various sitting posture recognition models, the sitting posture feature information of various body types of people sitting on the sofa in the test sets of various sitting posture recognition models is obtained, and the information is substituted into the recognition models of various sitting postures, and the sitting posture corresponding to the sitting posture feature information of various body types of people sitting on the sofa obtained based on the sitting posture recognition model is recorded as the predicted sitting posture of various body types of people sitting on the sofa, thereby obtaining the predicted sitting posture of various body types of people sitting on the sofa in the test sets of various sitting posture recognition models.

[0042] The predicted sitting postures of people of different body types sitting on the sofa in the test sets of various sitting posture recognition models are compared with their actual sitting postures. If the predicted sitting posture of a person of a certain body type sitting on the sofa is consistent with its actual sitting posture, the sitting posture of the person of this body type is recorded as the correctly recognized sitting posture of the person. The number of correctly recognized sitting postures of the person in the test sets of various sitting posture recognition models is counted, and the ratio between the number of correctly recognized sitting postures of the person in the test sets of various sitting posture recognition models and the total number of sitting postures of the person is further obtained, which is recorded as the recognition accuracy rate of various sitting posture recognition models.

[0043] Furthermore, the specific working process of the sitting posture recognition model training optimization module also includes: comparing the recognition accuracy of various sitting posture recognition models with a preset recognition accuracy threshold. If the recognition accuracy of a certain sitting posture recognition model is less than the preset recognition accuracy threshold, the accuracy of the sitting posture recognition model does not meet the requirements, and the statistical accuracy of various sitting posture recognition models does not meet the requirements.

[0044] As a preferred solution, the recognition accuracy threshold of the sitting posture recognition model is set according to design requirements. In a specific embodiment, the recognition accuracy threshold of the sitting posture recognition model is 90%, 92% or 95%.

[0045] The test sets of various sitting posture recognition models whose accuracy does not meet the requirements are merged with the training sets to obtain new training sets of various sitting posture recognition models whose accuracy does not meet the requirements; based on the new training sets of various sitting posture recognition models whose accuracy does not meet the requirements, the various sitting posture recognition models that do not meet the requirements are retrained, and the recognition models of various sitting postures that do not meet the requirements are optimized.

[0046] It should be noted that the present invention obtains the pressure distribution, weight distribution, contact area distribution, and back inclination angle of people of various body shapes sitting on the sofa in various sitting postures, trains recognition models for various sitting postures, determines whether their accuracy meets the requirements and performs optimization, analyzes the sitting posture recognition model from various aspects of sitting posture characteristics, and then improves the accuracy of the sitting posture recognition model, thereby improving the accuracy of the user sitting posture recognition results based on the sitting posture recognition model.

[0047] The user sitting posture matching and recognition module is used to collect the sitting posture feature information of the user when sitting on the sofa, combine the recognition models of various sitting postures, analyze the matching degree between the user's sitting posture feature information and various sitting postures, and further recognize the user's sitting posture.

[0048] Furthermore, the specific working process of using the user sitting posture matching and recognition module includes: dividing the areas to which the various components of the sofa belong according to preset principles into a grid format to obtain sub-areas in the various components of the sofa, wherein the various components of the sofa include a headrest, a backrest, a lumbar support, a seat cushion, a leg pad, a footrest, and an armrest.

[0049] By installing pressure detection sensors in each sub-area of ​​each component of the sofa, the pressure detected in each sub-area of ​​each component of the sofa when the user sits on the sofa is obtained, and the relative proportion of the pressure of each sub-area of ​​each component of the sofa when the user sits on the sofa is analyzed to obtain the pressure distribution when the user sits on the sofa.

[0050] As a preferred solution, the relative proportion of pressure in each sub-area of ​​each component of the sofa when the user sits on the sofa is analyzed. The specific method is: obtain the ratio of the pressure detected in each sub-area of ​​each component of the sofa when the user sits on the sofa to the cumulative pressure detected in the sub-area of ​​its component, and record it as the relative proportion of pressure in each sub-area of ​​each component of the sofa when the user sits on the sofa.

[0051] As a preferred solution, the pressure detection sensor may be a thin film pressure sensor, a piezoresistive sensor, etc.

[0052] By installing weight detection sensors in each sub-area of ​​each component of the sofa, the weight detected in each sub-area of ​​each component of the sofa when the user sits on the sofa is obtained, and the relative proportion of the weight of each sub-area of ​​each component of the sofa when the user sits on the sofa is analyzed to obtain the weight distribution of the user when sitting on the sofa.

[0053] As a preferred solution, the relative proportion of the weight of each sub-area in each component of the sofa when the user sits on the sofa is analyzed. The specific method is: obtain the ratio of the weight detected in each sub-area in each component of the sofa when the user sits on the sofa to the cumulative weight detected in the sub-area of ​​its component, and record it as the relative proportion of the weight of each sub-area in each component of the sofa when the user sits on the sofa.

[0054] As a preferred solution, the weight detection sensor can be a strain gauge weighing sensor, a capacitive weighing sensor, an inductive weighing sensor, a piezoresistive weighing sensor, a piezoelectric weighing sensor, etc.

[0055] In another specific implementation, the weight detected by each sub-area in each component of the sofa when the user sits on the sofa can be converted by using the pressure detected by each sub-area in each component of the sofa when the user sits on the sofa. The weight detected by the sub-area is obtained according to the pressure detected by the sub-area, and the specific method is: according to the pressure detected by the sub-area, the pressure detected in the vertical direction of the sub-area is obtained, and it is recorded as the gravity detected by the sub-area, and the weight detected by the sub-area is obtained according to the relationship between gravity and weight G=mg, (g is the acceleration of gravity), wherein, if the pressure acts vertically on the pressure detection sensor, the pressure detected by the sub-area is the gravity detected by the sub-area, otherwise the force is decomposed to obtain the gravity detected by the sub-area.

[0056] By installing contact area detection sensors in each sub-area of ​​each component of the sofa, the human body contact area detected in each sub-area of ​​each component of the sofa when the user sits on the sofa is obtained, and the relative proportion of the contact area of ​​each sub-area in each component of the sofa when the user sits on the sofa is analyzed to obtain the contact area distribution when the user sits on the sofa.

[0057] As a preferred solution, the relative proportion of the contact area of ​​each sub-area in each component of the sofa when the user sits on the sofa is analyzed. The specific method is: obtain the ratio of the human body contact area detected in each sub-area in each component of the sofa when the user sits on the sofa to the cumulative human body contact area detected by the sub-area of ​​its component, and record it as the relative proportion of the contact area of ​​each sub-area in each component of the sofa when the user sits on the sofa.

[0058] As a preferred solution, the contact area detection sensor can be a pressure sensor array, a capacitive sensor, a resistive sensor, an optical sensor, an ultrasonic sensor, etc.

[0059] The tilt angle of the back of the user when sitting on the sofa is obtained through the tilt angle sensor installed at the sofa back.

[0060] As a preferred solution, the tilt angle sensor can also be installed on the surface of the sofa. The tilt angle sensor determines the tilt angle of the back of the user when sitting on the sofa by calculating the change in the distance between the user's back and the sofa backrest.

[0061] It should be noted that the tilt angle sensor detects the change in the distance between the user's back and the sofa backrest to analyze the back tilt angle of the user when sitting on the sofa. The technology involved in the tilt angle sensor is an existing relatively mature technology and will not be elaborated here.

[0062] As a preferred solution, the method for obtaining the sitting posture characteristic information of the user when sitting on the sofa is the same as the method for obtaining the sitting posture characteristic information of people of various body shapes sitting on the sofa in various sitting postures.

[0063] Furthermore, the specific working process of using the user sitting posture matching recognition module also includes: S1: according to the recognition models of various sitting postures, obtaining the pressure distribution, weight distribution, contact area distribution, and back tilt angle corresponding to various sitting postures.

[0064] S2: The relative proportion of pressure in each sub-area of ​​each component of the sofa when the user sits on the sofa is recorded as i represents the number of the i-th component of the sofa, i = 1, 2, ..., n, n = 7, j represents the number of the j-th sub-area, j = 1, 2, ..., m, and the relative proportion of the pressure of each sub-area in each component of the sofa corresponding to various sitting postures is recorded as a represents the number of the ath sitting posture, a=1,2,...,b.

[0065] By analyzing the formula Obtain the matching factors of user sitting posture feature information and various sitting postures based on pressure distribution Wherein, e represents a natural constant, and Δk represents a preset threshold value of the deviation of the relative pressure ratio.

[0066] S3: Compare the weight distribution of the user when sitting on the sofa with the weight distribution corresponding to various sitting postures, and analyze the matching factors between the user's sitting posture feature information based on the weight distribution and the various sitting postures.

[0067] As a preferred solution, the method of analyzing the matching factors of the user's sitting posture characteristic information based on weight distribution and various sitting postures has the same principle as the method of analyzing the matching factors of the user's sitting posture characteristic information based on pressure distribution and various sitting postures.

[0068] S4: comparing the contact area distribution when the user sits on the sofa with the contact area distribution corresponding to various sitting postures, and analyzing the matching factors between the user's sitting posture feature information based on the contact area distribution and the various sitting postures.

[0069] As a preferred solution, the method of analyzing the matching factors of the user's sitting posture characteristic information based on the contact area distribution and various sitting postures has the same principle as the method of analyzing the matching factors of the user's sitting posture characteristic information based on the pressure distribution and various sitting postures.

[0070] S5: The back tilt angle when the user sits on the sofa and the back tilt angle corresponding to various sitting postures are recorded as θ, By analyzing the formula Obtain the matching factors of the user's sitting posture feature information and various sitting postures based on the back tilt angle in

[0071] S6: Calculate the weighted average of the matching factors between the user's sitting posture characteristic information and various sitting postures based on pressure distribution, weight distribution, contact area distribution, and back tilt angle to obtain the matching degree between the user's sitting posture characteristic information and various sitting postures.

[0072] As a preferred solution, the weights of the matching factors of the user's sitting posture characteristic information and various sitting postures based on pressure distribution, weight distribution, contact area distribution, and back tilt angle are set values, and the cumulative sum is 1. In a specific embodiment, the weights of the matching factors of the user's sitting posture characteristic information and various sitting postures based on pressure distribution, weight distribution, contact area distribution, and back tilt angle are 0.3, 0.2, 0.2, and 0.3, respectively.

[0073] Furthermore, the specific working process of the user sitting posture matching and identifying module also includes: comparing the user sitting posture feature information with the matching degrees of various sitting postures, and taking the sitting posture corresponding to the maximum matching degree as the sitting posture of the user.

[0074] As a preferred solution, the user's sitting posture feature information and the corresponding sitting posture are added to the training set of the sitting posture recognition model, and then the training set of the sitting posture recognition model is updated to further improve the accuracy of the sitting posture recognition model.

[0075] The module for reminding the user of poor sitting posture is used to compare the sitting posture of the user with a set of bad sitting postures, determine whether the sitting posture of the user is poor, and give a reminder.

[0076] Furthermore, the specific working process of the module for reminding users of bad sitting posture is as follows: extracting a bad sitting posture set stored in a database, comparing the sitting posture of the user with the bad sitting posture set; if the sitting posture of the user is consistent with a sitting posture in the bad sitting posture set, the user's sitting posture is bad, and the user is reminded.

[0077] As a preferred solution, bad sitting posture includes but is not limited to hunching over and hunching back.

[0078] As a preferred solution, the user can be reminded to adjust the sitting posture through vibration reminders, voice prompts, APP prompts, screen displays, shaking modes, alarms, etc.

[0079] The user sedentary reminder module is used to obtain the length of time the user maintains the same sitting posture and the cumulative length of time the user uses the sofa once, determine whether the user has been sedentary for a long time, and give a reminder.

[0080] Furthermore, the specific working process of the user sedentary reminder module is: obtaining the time the user maintains the same sitting posture and the cumulative time the user uses the sofa in a single time through the counter installed on the sofa, extracting the sedentary time threshold stored in the database, if the time the user maintains the same sitting posture and the cumulative time the user uses the sofa in a single time are both less than or equal to the sedentary time threshold, the user is not sedentary, otherwise, the user is sedentary, and the user is reminded.

[0081] As a preferred solution, the process from the user sitting on the sofa to the user getting up and leaving the sofa is recorded as the user using the sofa once.

[0082] It should be noted that the present invention collects the sitting posture feature information of the user when sitting on the sofa, combines it with various sitting posture recognition models, identifies the user's sitting posture, further determines whether the user's sitting posture is not good and whether the user has been sitting for a long time, and gives a reminder, which helps the user maintain good sitting habits, prevent physical diseases caused by long-term sitting, and thus improve the user's comfort and health level.

[0083] The sofa posture automatic control module is used to obtain the startup requirements and working modes of the sofa, wherein the working modes include default mode, custom mode, and normal mode, further analyze the target posture parameters of the sofa, and feed them back to the control end of the sofa, thereby controlling the posture of the sofa.

[0084] Further, see Figure 3As shown, the specific working process of the sofa posture automatic control module is: D1: obtaining the startup requirements and working mode of the sofa by using the user's mobile terminal, wherein the working mode includes a default mode, a custom mode, and a normal mode.

[0085] As a preferred solution, the user's mobile terminal includes but is not limited to: mobile phones, tablets, computers and other smart devices.

[0086] As a preferred solution, the user's mobile terminal is connected to the control end of the sofa through Wi-Fi, Bluetooth, Zigbee, etc.

[0087] D2: When the working mode of the sofa is the default mode, the posture parameters of the sofa in the default mode stored in the database are extracted and recorded as the target posture parameters of the sofa, where the posture parameters include the headrest inclination angle, backrest inclination angle, lumbar support inclination angle, seat cushion inclination angle, seat cushion height, leg pad inclination angle, leg pad extension length, foot rest inclination angle, foot rest extension length, and armrest height.

[0088] D3: When the working mode of the sofa is the custom mode, the posture parameters of the sofa input by the user's mobile terminal are obtained and recorded as the target posture parameters of the sofa.

[0089] D4: When the working mode of the sofa is the normal mode, the posture parameters of the sofa used by the user in history are obtained through the data storage module of the sofa, and are recorded as the target posture parameters of the sofa.

[0090] As a preferred solution, the posture parameters of the sofa that is historically used refer to the posture parameters of the sofa that is most frequently used historically.

[0091] D5: Feedback the target posture parameters of the sofa to the control end of the sofa, and then adjust the posture of the sofa.

[0092] As a preferred solution, the sofa contains multiple motor drive modules for controlling the electric adjustment functions of the sofa, such as lifting, extending, etc.

[0093] It should be noted that the present invention obtains the startup requirements and working mode of the sofa, further analyzes the target posture parameters of the sofa, and feeds them back to the control end of the sofa, thereby regulating the posture of the sofa, which is conducive to the smart sofa to achieve highly customized and personalized services, thereby providing users with the best usage experience and comfort.

[0094] The database is used to store a set of bad sitting postures, a threshold of long-sitting time, and posture parameters of the sofa in a default mode.

[0095] The above contents are merely examples and explanations of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. An intelligent sofa control system based on network technology, characterized in that: include: Sitting posture recognition model training and optimization module: used to obtain the sitting posture characteristic information of people of various body types sitting on the sofa in various sitting postures through virtual simulation technology, where the sitting posture characteristic information includes pressure distribution, weight distribution, contact area distribution, and back tilt angle, train various sitting posture recognition models, further determine whether their accuracy meets the requirements, and optimize the various sitting posture recognition models that do not meet the requirements; User sitting posture matching and recognition module: used to collect the user's sitting posture feature information when sitting on the sofa, combine various sitting posture recognition models, analyze the matching degree between the user's sitting posture feature information and various sitting postures, and further recognize the user's sitting posture; The user's bad sitting posture reminder module is used to compare the user's sitting posture with the bad sitting posture set, determine whether the user's sitting posture is bad, and remind the user; User long-term sitting reminder module: used to obtain the length of time the user maintains the same sitting posture and the cumulative length of time the user uses the sofa once, determine whether the user has been sitting for a long time, and give a reminder; Sofa posture automatic control module: used to obtain the startup requirements and working mode of the sofa, where the working mode includes default mode, custom mode, and normal mode, further analyze the target posture parameters of the sofa, and feed them back to the control end of the sofa, so as to control the posture of the sofa; Database: used to store bad sitting posture sets, sitting time thresholds, and sofa posture parameters in default mode.

2. According to claim 1, the intelligent sofa control system based on network technology is characterized in that: The specific working process of the sitting posture recognition model training optimization module includes: Taking height and weight as variables of human body shape, the variables of human body shape are set and adjusted multiple times according to the preset principles to obtain the body shape of each person; Using virtual simulation technology to simulate people of various body types sitting on the sofa in various sitting postures, and obtaining sitting posture characteristic information of people of various body types sitting on the sofa in various sitting postures; According to the sitting posture feature information of people of various body types sitting on the sofa in various sitting postures, the sitting posture feature information of people of various body types sitting on the sofa corresponding to the various sitting postures is classified and counted, and data sets of various sitting posture recognition models are constructed. The data sets of various sitting posture recognition models are divided according to the preset ratio between the training set and the test set to obtain the training set and the test set of various sitting posture recognition models; Based on the training sets of various sitting posture recognition models, machine learning algorithms are used to analyze the pressure distribution, weight distribution, contact area distribution, and back tilt angle corresponding to various sitting postures, and to train recognition models for various sitting postures.

3. The intelligent sofa control system based on network technology according to claim 2 is characterized in that: The specific working process of the sitting posture recognition model training optimization module also includes: According to the test sets of various sitting posture recognition models, the sitting posture feature information of the human bodies of various body types when sitting on the sofa in the test sets of various sitting posture recognition models is obtained, and the information is substituted into the recognition models of various sitting postures, and the sitting posture corresponding to the sitting posture feature information of the human bodies of various body types when sitting on the sofa obtained based on the sitting posture recognition models is recorded as the predicted sitting posture of the human bodies of various body types sitting on the sofa, thereby obtaining the predicted sitting posture of the human bodies of various body types sitting on the sofa in the test sets of various sitting posture recognition models; The predicted sitting postures of people of different body types sitting on the sofa in the test sets of various sitting posture recognition models are compared with their actual sitting postures. If the predicted sitting posture of a certain body type sitting on the sofa is consistent with its actual sitting posture, the sitting posture of this body type is recorded as the correctly recognized sitting posture of the human body. The number of correctly recognized sitting postures of the human body in the test sets of various sitting posture recognition models is counted, and the ratio between the number of correctly recognized sitting postures of the human body and the total number of sitting postures of the human body in the test sets of various sitting posture recognition models is further obtained, which is recorded as the recognition accuracy rate of various sitting posture recognition models.

4. The intelligent sofa control system based on network technology according to claim 3 is characterized in that: The specific working process of the sitting posture recognition model training optimization module also includes: Compare the recognition accuracy of various sitting posture recognition models with the preset recognition accuracy threshold. If the recognition accuracy of a certain sitting posture recognition model is less than the preset recognition accuracy threshold, the accuracy of the sitting posture recognition model does not meet the requirements, and the statistical accuracy of various sitting posture recognition models does not meet the requirements; The test sets of various sitting posture recognition models whose accuracy does not meet the requirements are merged with the training sets to obtain new training sets of various sitting posture recognition models whose accuracy does not meet the requirements; based on the new training sets of various sitting posture recognition models whose accuracy does not meet the requirements, the various sitting posture recognition models that do not meet the requirements are retrained, and the recognition models of various sitting postures that do not meet the requirements are optimized.

5. The intelligent sofa control system based on network technology according to claim 1 is characterized in that: The specific working process of using the user sitting posture matching and recognition module includes: According to the preset principle, the areas to which the various components of the sofa belong are divided into grids to obtain the sub-areas of the various components of the sofa, wherein the various components of the sofa include a headrest, a backrest, a lumbar support, a seat cushion, a leg pad, a footrest, and an armrest; The pressure detected in each sub-area of ​​each component of the sofa when the user sits on the sofa is obtained by installing a pressure detection sensor in each sub-area of ​​each component of the sofa, and the relative proportion of the pressure in each sub-area of ​​each component of the sofa when the user sits on the sofa is analyzed to obtain the pressure distribution when the user sits on the sofa; By installing weight detection sensors in each sub-area of ​​each component of the sofa, the weight detected in each sub-area of ​​each component of the sofa when the user sits on the sofa is obtained, and the relative proportion of the weight of each sub-area of ​​each component of the sofa when the user sits on the sofa is analyzed to obtain the weight distribution of the user when sitting on the sofa; By installing contact area detection sensors in each sub-area of ​​each component of the sofa, the human body contact area detected in each sub-area of ​​each component of the sofa when the user sits on the sofa is obtained, and the relative proportion of the contact area of ​​each sub-area in each component of the sofa when the user sits on the sofa is analyzed to obtain the contact area distribution when the user sits on the sofa; The tilt angle of the back of the user when sitting on the sofa is obtained through the tilt angle sensor installed at the sofa back.

6. The intelligent sofa control system based on network technology according to claim 5 is characterized in that: The specific working process of using the user sitting posture matching and recognition module also includes: S1: According to the recognition models of various sitting postures, the pressure distribution, weight distribution, contact area distribution and back tilt angle corresponding to various sitting postures are obtained; S2: The relative proportion of pressure in each sub-area of ​​each component of the sofa when the user sits on the sofa is recorded as i represents the number of the i-th component of the sofa, i = 1, 2, ..., n, n = 7, j represents the number of the j-th sub-area, j = 1, 2, ..., m, and the relative proportion of the pressure of each sub-area in each component of the sofa corresponding to various sitting postures is recorded as a represents the number of the a-th sitting posture, a=1,2,...,b; By analyzing the formula Obtain the matching factors of user sitting posture feature information and various sitting postures based on pressure distribution Wherein, e represents a natural constant, and Δk represents a preset threshold value of deviation of the relative pressure ratio; S3: comparing the weight distribution of the user when sitting on the sofa with the weight distribution corresponding to various sitting postures, and analyzing the matching factors between the user's sitting posture feature information based on the weight distribution and various sitting postures; S4: comparing the contact area distribution when the user sits on the sofa with the contact area distribution corresponding to various sitting postures, and analyzing the matching factors between the user's sitting posture feature information based on the contact area distribution and the various sitting postures; S5: The back tilt angle when the user sits on the sofa and the back tilt angle corresponding to various sitting postures are recorded as θ, By analyzing the formula Obtain the matching factors of the user's sitting posture feature information and various sitting postures based on the back tilt angle in S6: Calculate the weighted average of the matching factors between the user's sitting posture characteristic information and various sitting postures based on pressure distribution, weight distribution, contact area distribution, and back tilt angle to obtain the matching degree between the user's sitting posture characteristic information and various sitting postures.

7. The intelligent sofa control system based on network technology according to claim 6 is characterized by: The specific working process of using the user sitting posture matching and recognition module also includes: The user's sitting posture feature information is compared with the matching degrees of various sitting postures, and the sitting posture corresponding to the maximum matching degree is used as the user's sitting posture.

8. The intelligent sofa control system based on network technology according to claim 1 is characterized by: The specific working process of the module for reminding users of poor sitting posture is as follows: A bad sitting posture set stored in the database is extracted, and the user's sitting posture is compared with the bad sitting posture set. If the user's sitting posture is consistent with a sitting posture in the bad sitting posture set, the user's sitting posture is bad, and the user is reminded.

9. The intelligent sofa control system based on network technology according to claim 1 is characterized by: The specific working process of using the user sedentary reminder module is as follows: The length of time the user maintains the same sitting posture and the cumulative length of time the user uses the sofa in a single session are obtained through the counter installed on the sofa, and the sedentary time threshold stored in the database is extracted. If the length of time the user maintains the same sitting posture and the cumulative length of time the user uses the sofa in a single session are both less than or equal to the sedentary time threshold, the user is not sedentary; otherwise, the user is sedentary and the user is reminded.

10. The intelligent sofa control system based on network technology according to claim 1, characterized in that: The specific working process of the sofa posture automatic control module is as follows: D1: Obtaining the startup requirements and working mode of the sofa by using the user's mobile terminal, where the working mode includes a default mode, a custom mode, and a normal mode; D2: When the working mode of the sofa is the default mode, the posture parameters of the sofa in the default mode stored in the database are extracted and recorded as the target posture parameters of the sofa, wherein the posture parameters include the tilt angle of the headrest, the tilt angle of the backrest, the tilt angle of the lumbar support, the tilt angle of the seat cushion, the seat cushion height, the tilt angle of the leg pad, the extended length of the leg pad, the tilt angle of the foot rest, the extended length of the foot rest, and the armrest height; D3: When the working mode of the sofa is the custom mode, the posture parameters of the sofa input by the user's mobile terminal are obtained and recorded as the target posture parameters of the sofa; D4: When the working mode of the sofa is the normal mode, the posture parameters of the sofa used by the user in history are obtained through the data storage module of the sofa, and the parameters are recorded as the target posture parameters of the sofa; D5: Feedback the target posture parameters of the sofa to the control end of the sofa, and then adjust the posture of the sofa.

Images