Intelligent self-adaptive mattress

Abstract

The invention relates to the technical field of smart home, and discloses an intelligent self-adaptive mattress which comprises a bottom layer and a surface layer, and an air bag spring control assembly is mounted in the bottom layer; the air bag spring control assembly comprises 360 cavities formed in the bottom layer, a spring is fixedly installed on the inner bottom wall of each cavity, and an air bag column is installed in each spring. According to the intelligent self-adaptive mattress, by arranging 360 air bag columns of the air bag spring control assembly for regional independent adjustment and combining pressure data collection of a gravity sensing contact grid and SVM algorithm sleeping posture recognition, a supporting curve can be dynamically adjusted according to the real-time sleeping posture of a user, millimeter-level precise supporting is achieved, meanwhile, quick response smaller than 200 ms is achieved through a PID control algorithm, and the intelligent self-adaptive mattress is suitable for being used in the field of sleep. The problem of non-uniform pressure distribution of a traditional mattress is solved; a multi-algorithm fusion learning optimization system is matched, an optimal support strategy is predicted based on seven-day sleep data, and the deep sleep quality is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart home, and specifically to a smart adaptive mattress. Background Art

[0002] A mattress is an item placed between the human body and the bed to ensure that consumers can obtain healthy and comfortable sleep. There are various mattress materials, and mattresses made of different materials can bring different sleep effects to people.

[0003] Currently, the smart mattresses on the market only achieve local telescopic adjustment, and cannot automatically adapt to the curve of the human body according to the user's current environment. In particular, they lack precise support for key parts such as the cervical spine and lumbar spine, resulting in uneven distribution of spinal pressure, which is likely to cause discomfort during sleep or morning aches and pains. At the same time, they cannot truly achieve a completely relaxed state when people rest, nor can they detect the various physical states of the user in real time, reducing comfort and practicality. Therefore, a smart adaptive mattress is proposed to solve the above-mentioned problems. Summary of the Invention

[0004] (1) Technical Problems to be Solved

[0005] In view of the deficiencies of the prior art, the present invention provides a smart adaptive mattress, which has the advantages of strong adaptability, etc., and solves the problems that the smart mattresses on the market only achieve local telescopic adjustment, cannot automatically adapt to the curve of the human body according to the user's current environment, especially lack precise support for key parts such as the cervical spine and lumbar spine, resulting in uneven distribution of spinal pressure and being likely to cause discomfort during sleep or morning aches and pains.

[0006] (2) Technical Solutions

[0007] To achieve the above high practicality purpose, the present invention provides the following technical solutions: A smart adaptive mattress includes a bottom layer and a surface layer, and an airbag spring control component is installed inside the bottom layer;

[0008] The airbag spring control component includes three hundred and sixty cavities opened inside the bottom layer. A spring is fixedly installed on the inner bottom wall of each cavity. An airbag column is installed inside each spring. A high-resilience sponge layer is fixedly installed on the top of each spring. An installation cavity is provided at the bed tail of the bottom layer, and an air pump is fixedly installed inside the installation cavity;

[0009] A gravity sensing contact grid is laid on the top of each high-resilience sponge layer. A control system, a central processing unit and a Bluetooth module are fixedly installed inside the installation cavity. An alarm is installed on one side near the bed tail inside the bottom layer.

[0010] Preferably, a heat dissipation component is installed inside the installation cavity. A heat conduction copper plate and heat dissipation fins are installed inside the installation cavity. The surfaces of the air extraction pump, the control system, and the central processing unit are all in contact with one side of the heat conduction copper plate. The cold end of the heat dissipation fins is in contact with the surface of the heat conduction copper plate. The surface of the bottom layer is provided with no less than five heat dissipation holes, and the five heat dissipation holes communicate with the installation cavity.

[0011] Preferably, the bottom of each airbag column is connected to the air extraction pump through a pipeline, and a valve is installed on the surface of each pipeline.

[0012] Preferably, the bottom layer is designed with reinforced resin, and the surface layer is designed with a latex layer.

[0013] Preferably, the pressure data processing flow of the gravity sensing contact grid includes:

[0014] Pretreatment stage: Remove impulse noise through the median filtering algorithm, and the size of the filtering window is 5×5;

[0015] Feature extraction stage: Calculate 12 features such as the entropy value of pressure distribution, the offset of the centroid coordinate, and the change rate of the contact area;

[0016] Sleeping posture recognition stage: Adopt the support vector machine algorithm to distinguish three postures of lying flat, lying on the left side, and lying on the right side based on the Gaussian kernel function;

[0017] Among them, the central processing unit analyzes the pressure heat map based on the support vector machine algorithm of the Gaussian kernel function to identify the centroid coordinate and three postures of lying flat, lying on the left side, and lying on the right side.

[0018] Preferably, the gravity sensing contact grid uses the bicubic interpolation algorithm to generate a high-precision pressure heat map, and combines the Kalman filtering technology to realize real-time noise reduction processing of dynamic pressure data, ensuring that the system can accurately capture the user's turning action and respond quickly.

[0019] Preferably, the central processing unit adjusts the airbag column pressure through the PID control algorithm. The specific formula is:

[0020] Adjustment amount = Kp×deviation + Ki×integral + Kd×derivative, where Kp is the proportional coefficient, K is the integral coefficient, and Kd is the derivative coefficient;

[0021] The control system controls the air extraction pump to inflate and deflate the airbag column through the solenoid valve group, and the single-time adjustment response time < 200ms;

[0022] The high-resilience sponge layer on the top of the spring forms a spring matrix plate through an elastic connection material, and cooperates with the vertical expansion of the airbag column to achieve regional support.

[0023] Preferably, it further includes a learning and optimization system, where the learning and optimization system uses a machine learning model to predict the best support strategy. The input of the model is the night turning frequency and pressure distribution data for 7 consecutive days, and the output is the target pressure values of 360 airbag columns;

[0024] The model is a two-layer LSTM network, with each layer containing 64 neurons. The activation function is tanh, trained by the mean squared error loss function, and the optimizer is RMSprop. After training, the model parameters are stored in the EEPROM of the central processing unit.

[0025] Preferably, when the pressure heat map data is abnormal, an alarm is sent to the mobile phone APP through the Bluetooth module, and the local alarm of the alarm is triggered.

[0026] Beneficial effects

[0027] Compared with the prior art, the present invention provides an intelligent adaptive mattress, which has the following beneficial effects:

[0028] 1. For this intelligent adaptive mattress, through the independent adjustment of 360 airbag columns in the airbag spring control component by region, combined with the pressure data collection of the gravity sensing contact grid and the SVM algorithm for sleeping posture recognition, the support curve can be dynamically adjusted according to the real-time sleeping posture of the user, achieving millimeter-level precise support. At the same time, through the PID control algorithm, a fast response of <200ms is achieved, solving the problem of uneven pressure distribution of traditional mattresses; equipped with a learning and optimization system that integrates multiple algorithms, predicting the best support strategy based on 7-day sleep data, improving the quality of deep sleep, and being able to monitor abnormal pressure heat maps in real time, warning of the body state, and realizing automated health management.

[0029] 2. For this intelligent adaptive mattress, the heat dissipation component in the installation cavity is combined with a heat conduction copper plate and heat dissipation fins to quickly export the heat of core components such as the air pump and control system, and dissipate heat through the heat dissipation holes in communication with the outside air, avoiding equipment performance attenuation or failure caused by high temperature, and ensuring the stability and service life of the mattress under high-frequency use; at the same time, the dual warning mechanism of the local alarm and the Bluetooth module can respond to abnormal states in a timely manner, improving the safety of the sleep scenario, especially suitable for demand scenarios such as the elderly living alone and medical care. Description of the drawings

[0030] Figure 1 It is a three-dimensional structure diagram of the intelligent adaptive mattress of the present invention;

[0031] Figure 2 It is a top cross-sectional view of the structure of the intelligent adaptive mattress of the present invention;

[0032] Figure 3 It is a partial cross-sectional view of the structure of the airbag spring control component of the present invention;

[0033] Figure 4 Top view of the intelligent adaptive mattress structure of the present invention;

[0034] Figure 5 Schematic diagram of the intelligent adaptive mattress structure framework of the present invention;

[0035] Figure 6 Schematic diagram of the intelligent adaptive mattress structure circuit of the present invention.

[0036] In the figure: 101, bottom layer; 102, surface layer; 2, airbag spring control component; 201, cavity; 202, spring; 203, airbag column; 204, high-resilience sponge layer; 205, installation cavity; 206, air pump; 3, gravity sensing contact grid; 4, control system; 5, central processing unit; 6, alarm; 7, Bluetooth module; 8, heat dissipation component; 801, heat-conducting copper plate; 802, heat dissipation fins; 803, heat dissipation holes. Specific embodiments

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0038] Embodiment 1

[0039] Please refer to Figures 1-6 , an intelligent adaptive mattress, including a bottom layer 101 and a surface layer 102. The bottom layer 101 is designed with reinforced resin, and the surface layer 102 is designed with a latex layer. An airbag spring control component 2 is installed inside the bottom layer 101;

[0040] The airbag spring control component 2 includes three hundred and sixty cavities 201 opened inside the bottom layer 101. A spring 202 is fixedly installed on the inner bottom wall of each cavity 201. An airbag column 203 is installed inside each spring 202. A high-resilience sponge layer 204 is fixedly installed on the top of each spring 202. An installation cavity 205 is provided at the end of the bottom layer 101 of the bed, and an air pump 206 is fixedly installed inside the installation cavity 205;

[0041] The bottom of each airbag column 203 is connected to the air pump 206 through a pipeline, and a valve is installed on the surface of each pipeline.

[0042] The pressure data processing process of the gravity sensing contact grid 3 includes:

[0043] Preprocessing stage: Remove pulse noise through the median filtering algorithm, and the filter window size is 5×5;

[0044] Feature extraction stage: Calculate 12 features such as pressure distribution entropy value, center of gravity coordinate offset, contact area change rate, etc.;

[0045] Sleep posture recognition stage: Adopt the support vector machine (SVM) algorithm, and based on the Gaussian kernel function, distinguish three postures of lying flat, left lateral lying, and right lateral lying;

[0046] Among them, the central processing unit 5 analyzes the pressure heat map based on the support vector machine (SVM) algorithm with the Gaussian kernel function to identify the center of gravity coordinates and three postures of lying flat, left lateral lying, and right lateral lying.

[0047] The gravity sensing contact grid 3 uses the bicubic interpolation algorithm to generate a high-precision pressure heat map, and combines the Kalman filtering technology to realize real-time noise reduction processing of dynamic pressure data, ensuring that the system can accurately capture the user's turning actions and respond quickly.

[0048] The central processing unit 5 adjusts the pressure of the airbag column 203 through the PID control algorithm. The specific formula is:

[0049] Adjustment amount = Kp × deviation + Ki × integral + Kd × differential, where Kp is the proportional coefficient, K is the integral coefficient, and Kd is the differential coefficient;

[0050] The control system 4 controls the air extraction pump 206 to inflate and deflate the airbag column 203 through the solenoid valve group, and the single adjustment response time < 200 ms;

[0051] The high-resilience sponge layer 204 on the top of the spring 202 forms a spring matrix plate through the elastic connection material, and cooperates with the airbag column 203 to vertically expand and restrict the horizontal expansion to achieve zonal support.

[0052] It also includes a learning and optimization system. Among them, the learning and optimization system uses a machine learning model to predict the best support strategy. The model input is the night turning frequency and pressure distribution data for 7 consecutive days, and the output is the target pressure values of 360 airbag columns 203;

[0053] The model is a two-layer LSTM network. Each layer contains 64 neurons. The activation function is tanh. It is trained through the mean square error (MSE) loss function, and the optimizer is RMSprop. After training, the model parameters are stored in the EEPROM of the central processing unit 5.

[0054] In this embodiment, in the pressure sensing stage: when the user lies down, the gravity is transmitted through the surface layer 102 to the high-resilience sponge layer 204, triggering the lower gravity sensing contact grid 3 to collect pressure data. After the pressure data is preprocessed and feature extracted, the central processing unit 5 uses the SVM algorithm to identify the sleep postures of lying flat, left lateral lying, right lateral lying and the center of gravity coordinates;

[0055] Dynamic adjustment stage: The central processing unit 5 calculates the target pressure value adjustment formula for the airbag column 203 according to the sleeping position and pressure distribution through the PID control algorithm: Adjustment = Kp × Deviation + Ki × Integral + Kd × Differential; The control system 4 controls the air extraction pump 206 to inflate and deflate the airbag column 203 through the solenoid valve group: When enhanced support is required, the air extraction pump 206 inflates the airbag column 203, and the airbag column 203 expands to lift the spring 202, enhancing the local support force;

[0056] When pressure reduction is needed, the air extraction pump 206 extracts air, the airbag column 203 contracts, and the spring 202 elastically returns, reducing the local pressure; The three hundred and sixty cavities 201 are divided into thirty groups with 12 in each group. Each group is connected to the air extraction pump 206 through an independent pipeline and valve, and the high-resilience sponge layer 204 on top of the spring 202 forms a spring matrix plate through elastic connection materials, and cooperates with the vertical expansion and horizontal expansion limitation of the airbag column 203 to achieve zonal support;

[0057] Learning and optimization stage

[0058] The learning and optimization system analyzes the data of the turning frequency and pressure distribution for 7 consecutive days through a two-layer LSTM network model, predicts the best support strategy, and outputs the target pressure values of the 360 airbag columns 203 to achieve the dynamic optimization of the personalized support plan.

[0059] In this embodiment, the intelligent adaptive mattress can achieve the advantages of automatically and accurately adjusting the support according to the sleeping position, real-time monitoring of the body state, and warning of abnormalities by setting the airbag spring control component 2, the gravity sensing contact grid 3, and the multi-algorithm fusion.

[0060] Embodiment 2

[0061] Please refer to Figures 1-6 , the intelligent adaptive mattress, where a gravity sensing contact grid 3 is laid on top of each high-resilience sponge layer 204, a control system 4, a central processing unit 5, and a Bluetooth module 7 are fixedly installed inside the installation cavity 205, and an alarm 6 is installed inside the bottom layer 101 near the end of the bed.

[0062] When the pressure heat map data shows abnormalities such as no pressure change for a long time or the center of gravity deviation exceeding the safety threshold, an alarm is sent to the mobile APP through the Bluetooth module 7, and the local alarm of the alarm 6 is triggered.

[0063] In this embodiment, the intelligent adaptive mattress collects pressure data in real time through the gravity sensing contact grid 3 and generates a pressure heat map. The central processing unit 5 analyzes the data to identify abnormal states such as no pressure change for a long time or the center of gravity deviation exceeding the limit, and then sends an alarm to the mobile APP through the Bluetooth module 7, and at the same time triggers the alarm 6 to give a local warning.

[0064] This design can detect the abnormal conditions of users in a timely manner, achieve dual local and remote warnings, improve sleep safety, and is especially suitable for scenarios such as the elderly living alone, medical care, etc., making users and their families feel more at ease.

[0065] A heat dissipation component 8 is installed inside the installation cavity 205. A heat conduction copper plate 801 and heat dissipation fins 802 are installed inside the installation cavity 205. The surfaces of the air extraction pump 206, the control system 4, and the central processing unit 5 are all in contact with one side of the heat conduction copper plate 801. The cold end of the heat dissipation fins 802 is in contact with the surface of the heat conduction copper plate 801. The surface of the bottom layer 101 is provided with no less than five heat dissipation holes 803, and the five heat dissipation holes 803 communicate with the installation cavity 205.

[0066] In this embodiment, the heat dissipation component 8 in the installation cavity 205 is attached to the air extraction pump 206, the control system 4, and the central processing unit 5 through the heat conduction copper plate 801, conducts the heat generated by the operation of the device to the heat dissipation fins 802, and then realizes heat dissipation through the air circulation between the heat dissipation holes 803 and the external air.

[0067] This design can quickly export the heat of electronic components and mechanical parts, avoid the decline of device performance or failures caused by high temperatures, ensure the stable operation of core components such as the control system 4 and the air pump in the mattress, extend the service life, and at the same time improve the use safety. Especially in the scenario of long-term high-frequency use, it can continuously maintain a reliable heat dissipation effect.

[0068] To sum up, this intelligent adaptive mattress, through the independent adjustment of 360 airbag columns 203 of the airbag spring control component 2 in different regions, combined with the pressure data collection of the gravity sensing contact grid 3 and the SVM algorithm for sleeping posture recognition, can dynamically adjust the support curve according to the user's real-time sleeping posture, achieve millimeter-level precise support, and at the same time achieve a fast response of <200ms through the PID control algorithm, solving the problem of uneven pressure distribution of traditional mattresses; equipped with a learning optimization system LSTM model that combines multiple algorithms, based on 7-day sleep data to predict the best support strategy, improve the quality of deep sleep, and can monitor the abnormality of the pressure heat map in real time, warning of physical conditions such as the risk of falling and no body movement for a long time, realizing automated health management.

[0069] Moreover, by setting the heat dissipation component 8 in the installation cavity 205 in combination with the heat conduction copper plate 801 and the heat dissipation fins 802, the heat of core components such as the air extraction pump 206 and the control system 4 is quickly exported, and heat dissipation is achieved through the air circulation between the heat dissipation holes 803 and the outside air, avoiding the attenuation of device performance or failures caused by high temperatures, ensuring the stability and service life of the mattress under high-frequency use; at the same time, the dual warning mechanism of the local alarm 6 and the Bluetooth module 7 can respond to abnormal states in a timely manner, improving the safety of the sleep scenario, and is especially suitable for demand scenarios such as the elderly living alone and medical care.

[0070] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising said element.

[0071] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The intelligent adaptive mattress includes a bottom layer (101) and a surface layer (102), and is characterized in that: An airbag spring control component (2) is installed inside the bottom layer (101). The airbag spring control component (2) includes 360 cavities (201) opened inside the bottom layer (101). A spring (202) is fixedly installed on the inner bottom wall of each cavity (201). An airbag column (203) is installed inside each spring (202). A high-resilience sponge layer (204) is fixedly installed on the top of each spring (202). An installation cavity (205) is arranged at the end of the bed of the bottom layer (101). An air pump (206) is fixedly installed inside the installation cavity (205). A gravity-sensing contact grid (3) is laid on the top of each high-resilience sponge layer (204). A control system (4), a central processing unit (5) and a Bluetooth module (7) are fixedly installed inside the installation cavity (205). An alarm (6) is installed inside the bottom layer (101) near one side of the end of the bed.

2. The intelligent adaptive mattress according to claim 1, characterized in that: A heat dissipation component (8) is installed inside the installation cavity (205). A heat-conducting copper plate (801) and heat dissipation fins (802) are installed inside the installation cavity (205). The surfaces of the air pump (206), the control system (4) and the central processing unit (5) are all in contact with one side of the heat-conducting copper plate (801). The cold end of the heat dissipation fin (802) is in contact with the surface of the heat-conducting copper plate (801). The surface of the bottom layer (101) is provided with no less than five heat dissipation holes (803), and the five heat dissipation holes (803) communicate with the installation cavity (205).

3. The intelligent adaptive mattress according to claim 1, wherein: The bottom of each airbag column (203) is connected to the air pump (206) through a pipeline, and a valve is installed on the surface of each pipeline.

4. The intelligent adaptive mattress according to claim 1, wherein: The bottom layer (101) is designed with reinforced resin, and the surface layer (102) is designed with a latex layer.

5. The intelligent adaptive mattress according to claim 1, wherein: The pressure data processing flow of the gravity-sensing contact grid (3) includes: Preprocessing stage: Remove pulse noise through the median filtering algorithm, and the size of the filtering window is 5×5. Feature extraction stage: Calculate 12 features such as the pressure distribution entropy value, the offset of the centroid coordinates, and the change rate of the contact area. Sleeping posture recognition stage: Adopt the support vector machine (SVM) algorithm, and distinguish three postures of lying flat, lying on the left side, and lying on the right side based on the Gaussian kernel function. Among them, the central processing unit (5) analyzes the pressure heat map based on the support vector machine (SVM) algorithm of the Gaussian kernel function, and identifies the centroid coordinates and three postures of lying flat, lying on the left side, and lying on the right side.

6. The intelligent adaptive mattress according to claim 1, wherein: The gravity-sensing contact grid (3) uses the bicubic interpolation algorithm to generate a high-precision pressure heat map, and combines the Kalman filtering technology to realize real-time noise reduction processing of dynamic pressure data, ensuring that the system can accurately capture the user's turning action and respond quickly.

7. The intelligent adaptive mattress according to claim 1, characterized in that: The central processing unit (5) adjusts the pressure of the airbag column (203) through the PID control algorithm. The specific formula is: Adjustment amount = Kp×Deviation + Ki×Integral + Kd×Differential, where Kp is the proportional coefficient, K is the integral coefficient, and Kd is the differential coefficient. The control system (4) controls the air extraction pump (206) to inflate and deflate the airbag column (203) through a solenoid valve group, and the single - adjustment response time < 200 ms; The high - resilience sponge layer (204) at the top of the spring (202) forms a spring matrix plate through an elastic connection material, and cooperates with the vertical expansion (restricting the lateral expansion) of the airbag column (203) to achieve zonal support.

8. The intelligent adaptive mattress according to claim 1, characterized in that: It also includes a learning and optimization system. The learning and optimization system uses a machine - learning model to predict the best support strategy. The model inputs are the night - time turning frequency and pressure distribution data for 7 consecutive days, and the output is the target pressure values of 360 airbag columns (203); The model is a two - layer LSTM network, with each layer containing 64 neurons. The activation function is tanh, trained through the mean - square error (MSE) loss function, and the optimizer is RMSprop. After training, the model parameters are stored in the EEPROM of the central processing unit (5).

9. The intelligent adaptive mattress according to claim 6, wherein: When the pressure heat - map data is abnormal (such as no pressure change for a long time, center - of - gravity deviation exceeding the safety threshold), an alarm is sent to the mobile phone APP through the Bluetooth module (7), and the local warning of the alarm (6) is triggered.

Images