Method for optimizing pressure distribution of intelligent mattress with multiple air pumps working cooperatively

Abstract

The invention discloses a pressure distribution optimization method for an intelligent mattress with multiple air pumps working cooperatively, and relates to the technical field of mattress pressure optimizing.The method comprises the steps that a plurality of independent air chamber areas in the intelligent mattress are determined, and a neighborhood air chamber pressure interference relation is established; performing pressure data collection on the target user, and performing contact surface analysis according to the pressure collection data to generate a user analysis posture; acquiring a preset spine alignment curve database; inputting the user analysis posture into the preset spine alignment curve database to obtain a target reference spine alignment curve; and carrying out pressure distribution optimization on the plurality of independent air chamber areas, and carrying out pressure distribution control. The technical problem that in the prior art, it is difficult for an intelligent mattress to accurately optimize pressure distribution according to the individual characteristics and the real-time posture of the user is solved, and the technical effects that accurate optimization control is conducted on the pressure distribution of the intelligent mattress according to the individual characteristics and the real-time posture of the user, and the sleep comfort of the user and the mattress adaptability are effectively improved are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of mattress pressure optimization, and in particular to a method for optimizing pressure distribution of an intelligent mattress with multiple air pumps working in coordination. Background Art

[0002] With the improvement of people's living standards and the increasing emphasis on healthy sleep, smart mattresses have gradually come into the public eye as a product that can provide a personalized sleeping experience. In the development of smart mattresses, although early products tried to introduce some simple adjustment functions, such as adjusting the overall air pressure of the mattress through a single air pump, this method could not achieve independent control of different areas of the mattress and could not provide differentiated support based on the specific needs of different parts of the user's body. Moreover, in the design of smart mattresses with multi-chamber structures, the problem of mutual interference between the pressures of the air chambers has always been a key factor restricting its performance improvement. Due to the lack of in-depth understanding and effective handling of the pressure interference relationship between neighboring air chambers, when the pressure of a certain air chamber changes, it will have an unpredictable impact on the adjacent air chambers, thereby destroying the stability and uniformity of the pressure distribution on the entire mattress surface, making it difficult for the mattress to provide users with the expected precise and comfortable support effect.

[0003] The existing technology has the technical problem that it is difficult for smart mattresses to accurately optimize pressure distribution according to the user's individual characteristics and real-time posture. Summary of the Invention

[0004] The present application provides a method for optimizing the pressure distribution of a smart mattress with multiple air pumps working in collaboration, which is used to solve the technical problem in the prior art that it is difficult to accurately optimize the pressure distribution of a smart mattress according to the user's individual characteristics and real-time posture.

[0005] In view of the above problems, the present application provides a method for optimizing pressure distribution of a smart mattress with multiple air pumps working in coordination, the method comprising: Determine multiple independent air chamber areas in the smart mattress, perform pressure interference analysis on adjacent air chambers, and establish a neighborhood air chamber pressure interference relationship; collect pressure data of the target user through the surface pressure sensor network of the smart mattress, perform contact surface analysis based on the pressure collection data, and generate a user analysis posture; obtain a preset spinal alignment curve database, wherein the preset spinal alignment curve database includes multiple postures of the target user and corresponding multiple reference spinal alignment curves; input the user analysis posture into the preset spinal alignment curve database, compare it with the multiple postures, and obtain a target reference spinal alignment curve; with the target reference spinal alignment curve as the target, optimize the pressure distribution of the multiple independent air chamber areas in combination with the neighborhood air chamber pressure interference relationship, generate multiple air chamber control parameters, and perform pressure distribution control.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The system identifies multiple independent air chamber regions within the smart mattress, performs pressure interference analysis on adjacent air chambers, and establishes pressure interference relationships between neighboring air chambers. Pressure data from the target user is collected through the smart mattress's surface pressure sensor network to generate a user-analyzed posture. A database of preset spinal alignment curves is obtained. The user-analyzed posture is input into the database and compared with multiple postures to obtain a target baseline spinal alignment curve. Pressure distribution is optimized for the multiple independent air chamber regions, generating multiple air chamber control parameters for pressure distribution control. This system achieves the technical effect of precisely optimizing and controlling the pressure distribution of the smart mattress based on the user's individual characteristics and real-time posture, effectively improving sleep comfort and mattress compatibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A flow chart of a method for optimizing pressure distribution in a smart mattress with multiple air pumps working in collaboration provided by an embodiment of the present application; Figure 2 A flow chart of establishing the pressure interference relationship of neighboring air chambers in the intelligent mattress pressure distribution optimization method with multiple air pumps working in collaboration provided in an embodiment of the present application. DETAILED DESCRIPTION

[0009] This application provides a method for optimizing the pressure distribution of a smart mattress with multiple air pumps working in collaboration, which is used to solve the technical problem in the existing technology that it is difficult to accurately optimize the pressure distribution of a smart mattress according to the user's individual characteristics and real-time posture.

[0010] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work are within the scope of protection of this application.

[0011] Examples, such as Figure 1 As shown, the present application provides a method for optimizing pressure distribution of an intelligent mattress with multiple air pumps working in coordination, the method comprising: Step S100: determining multiple independent air chamber areas in the smart mattress, performing pressure interference analysis on adjacent air chambers, and establishing a pressure interference relationship between adjacent air chambers.

[0012] Specifically, the internal structure of the mattress must first be carefully explored. Through physical structure analysis and electrical connection testing, multiple independent air chamber areas are accurately located and determined, each of which is equipped with an air pump and air bag. Subsequently, pressure interference analysis of adjacent air chambers is carried out, and experimental tests are performed using the control variable method. During the test, the pressure of the other air chambers is first fixed unchanged, and the pressure of one air chamber is gradually changed. A high-precision pressure sensor is used to monitor the pressure changes of the adjacent air chambers in real time to obtain multiple sets of pressure data. The data are then processed using a multivariate linear regression algorithm, with the pressure changes of the adjacent air chambers set as the dependent variable, and the pressure of the air chamber that actively changes the pressure and other related factors set as independent variables. The regression equation is fitted using the least squares method, and the coefficients in the equation can represent the pressure interference relationship. Based on the results of the regression equation, a mathematical model of the pressure interference relationship of neighboring air chambers is established, providing a scientific basis for the subsequent precise regulation of mattress pressure, thereby improving the user's sleep experience.

[0013] Step S200: collecting pressure data of a target user through the surface pressure sensor network of the smart mattress, performing contact surface analysis based on the pressure collection data, and generating a user analysis posture.

[0014] Specifically, a network of pressure sensors is meticulously arranged on the surface of the smart mattress, covering every key location. When the user lies on the mattress, the sensor network begins collecting pressure data in real time. During this process, the sensors record the pressure at each location at an extremely high frequency (e.g., dozens of times per second), generating a massive amount of raw pressure data points. Once this data is collected, the contact surface analysis phase begins. First, the raw pressure data is preprocessed. A filtering algorithm removes outliers caused by sensor noise or external interference to ensure data accuracy. Next, an interpolation algorithm is used to supplement any missing data, providing a more complete spatial distribution of the pressure data. Based on this processed data, the pressure distribution is visualized, presenting the pressure distribution on the mattress surface in an intuitive graphical form, clearly demonstrating pressure fluctuations and the density of the pressure distribution. The resulting pressure distribution map is then fed into an advanced connected region detection model. This model uses image segmentation techniques to divide the mattress surface into connected regions based on pressure differences. These regions roughly correspond to the different areas of contact between the user's body and the mattress. For example, areas of greater pressure may correspond to the user's torso, while areas of less pressure may be located on the limbs. Finally, the contact surface posture recognition results are compared with a pre-established posture matching list, which contains numerous corresponding contact surface posture samples and user posture samples obtained through extensive experiments and data analysis. By calculating the similarity between the recognition results and the samples in the list, such as using a cosine similarity algorithm or Euclidean distance algorithm, the best matching user posture sample is found, thereby generating an accurate user parsed posture, providing a key basis for subsequent mattress pressure optimization, and achieving the goal of providing users with personalized comfortable support.

[0015] Step S300: Acquire a preset spinal alignment curve database, wherein the preset spinal alignment curve database includes multiple postures of the target user and corresponding multiple reference spinal alignment curves.

[0016] Specifically, a key database of preset spinal alignment curves is obtained. This database is obtained through various methods. Firstly, it is obtained by leveraging existing mature technologies through the collection, analysis, and induction of a large amount of human spinal morphological data. For example, high-precision 3D scanning equipment is used to accurately scan the spinal morphology of people of different ages, genders, and body types in various standard sleeping positions (such as supine, side-lying, and prone), generating a massive amount of spinal point cloud data. This data is then processed using professional image processing and data analysis software to extract key characteristic points of the spine. Based on these characteristic points, corresponding spinal alignment curves are fitted and stored in the database as baseline spinal alignment curves. Secondly, experienced experts in relevant fields are invited to participate in the pre-setting. Based on ergonomic principles, medical research results, and long-term clinical experience, these experts make professional judgments and set the ideal alignment state of the human spine in different postures. Taking into account factors such as the human skeletal structure, muscle distribution, and physiological curvature, a scientifically sound baseline spinal alignment curve is developed for each posture. When building the database, the target user's various postures (such as lying flat, left side, right side, and half-sitting) are precisely associated and stored with corresponding multiple baseline spinal alignment curves. These postures cover all common positions a user may assume during sleep, ensuring that the database comprehensively and accurately reflects the standard spinal morphology under different postures. This provides a solid and reliable reference for subsequent smart mattresses to optimize pressure distribution based on the user's actual posture, thereby improving sleep quality.

[0017] Step S400: inputting the user's analyzed posture into the preset spinal alignment curve database, comparing it with the multiple postures, and obtaining a target reference spinal alignment curve.

[0018] Specifically, the user's interpreted posture data, collected and analyzed by a network of pressure sensors on the mattress surface, is input into a pre-set spinal alignment curve database. This database stores a variety of target user postures and their corresponding precise baseline spinal alignment curves. Key features of the user's interpreted posture, such as body contour shape, joint angle distribution, and pressure center of gravity location, are extracted to form a feature vector. This feature vector is then compared with the feature vectors of various postures stored in the database. During this comparison, the similarity between the two is calculated, using various established similarity calculation methods such as cosine similarity and Euclidean distance. A pre-set similarity threshold is defined. When the calculated similarity exceeds this threshold, the posture is considered to be highly matched with a posture in the database. The corresponding baseline spinal alignment curve is then obtained and designated as the target baseline spinal alignment curve. This target curve serves as a key reference for subsequent mattress pressure distribution optimization, ensuring that the mattress provides optimal support based on the user's actual posture, maintaining healthy spinal alignment and enhancing sleep comfort.

[0019] Step S500: Taking the target reference spinal alignment curve as the target, optimizing the pressure distribution of the plurality of independent air chamber regions in combination with the pressure interference relationship of the neighboring air chambers, generating a plurality of air chamber control parameters, and performing pressure distribution control.

[0020] Specifically, using the target baseline spinal alignment curve as the core guide and fully integrating the previously established neighboring air cell pressure interference relationships, a comprehensive and detailed pressure distribution optimization process is performed for multiple independent air cell regions within the mattress. First, based on the ideal human spine morphology and force distribution information contained in the target baseline spinal alignment curve, the ideal pressure range for each body part on the mattress is determined. For example, key areas supporting the spine should be applied with appropriately higher pressure to maintain the spine's natural curvature, while areas such as the limbs should be adapted to a relatively lower, comfortable pressure. Next, considering the neighboring air cell pressure interference relationships, due to the mutual influence between air cells, adjusting the pressure of one air cell will affect adjacent air cells. Therefore, when calculating the target pressure for each air cell, dynamic compensation based on the interference relationship is required. The neighboring air cell pressure interference coefficient is incorporated into the calculation to accurately calculate the actual control pressure that each independent air cell region should achieve, taking into account the interference. Based on these calculation results, multiple air cell control parameters are generated. These parameters clearly define key control instructions within each air cell, such as the output power of the air pump and the inflation or deflation volume of the airbag. Finally, based on these air chamber control parameters, the air pumps and air bags within the mattress's multiple independent air chambers are precisely controlled. The air pumps operate at a set power level, precisely adjusting the amount of air within the bags to achieve precise pressure distribution across the mattress's surface. Throughout this process, pressure changes are continuously monitored and provided with real-time feedback, allowing control parameters to be fine-tuned based on actual conditions. This ensures the mattress consistently provides comfortable and healthy pressure support that meets the user's target baseline spinal alignment curve, effectively enhancing the user's sleep experience.

[0021] In one possible implementation, Figure 2 As shown, step S100 also includes: Step S110: obtaining the distribution positions of the plurality of independent air chamber areas, combining the air chambers at adjacent positions according to the distribution positions, and generating a plurality of neighborhood air chamber combination results.

[0022] Step S120: performing a bidirectional pressure passive change test and analysis on the combination results of the plurality of neighboring air chambers to generate a plurality of passive pressure interference coefficients.

[0023] Step S130: generating the neighboring air chamber pressure interference relationship using the multiple passive pressure interference coefficients.

[0024] Specifically, specialized mattress internal structure detection technology is first used to accurately determine the distribution of multiple independent air cell regions within the mattress. This information includes the center coordinates, boundary range, and relative position of each cell in the mattress plane and vertically. Once the cell distribution is determined, the adjacent cell combinations are performed. Starting with the first cell, the algorithm searches for adjacent cells in all directions according to pre-defined proximity determination rules. For example, cells within a certain horizontal and vertical distance are considered adjacent. Once a cell is identified as an adjacent cell, the two cells are combined to form a neighborhood cell combination result. Because a cell may have multiple adjacent cells in different directions, it is repeatedly used for combinations. For example, if a cell has adjacent cells above, below, to the left, and to the right, it will be combined with each of these four cells in turn, producing four different combinations. This process continues for each cell, continuously searching and combining its adjacent cells until all cells have been combined, ultimately generating a comprehensive and accurate set of multiple neighborhood cell combination results. These combined results cover all possible pairs of air chambers in the mattress that may affect each other, laying a solid foundation for subsequent in-depth analysis of the pressure interference relationship between adjacent air chambers. This ensures that the interaction between air chambers can be fully considered during the mattress pressure distribution optimization process, thereby providing users with a more precise and comfortable support experience.

[0025] Based on the previously obtained results for multiple neighboring air cell combinations, bidirectional pressure passive variation testing and in-depth analysis were conducted between adjacent air cells, generating multiple passive pressure interference coefficients. For each neighboring air cell combination, consisting of a first air cell and a second air cell, testing was initially conducted in the first direction, with the first air cell set as the active interference source and the second air cell as the passive response chamber. Using a high-precision and precisely controllable pressure generator, the pressure within the first air cell was gradually varied. During this process, a highly sensitive pressure sensor accurately and in real time monitored the passive pressure variation within the second air cell. The corresponding second air cell pressure variation data for each first air cell pressure variation was recorded, forming a first-direction pressure passive variation test dataset. Following the first-direction test, testing was then conducted in the second direction, with the second air cell acting as the active interference source and the first air cell acting as the passive response chamber. The aforementioned pressure variation and monitoring process was repeated to obtain a second-direction pressure passive variation test dataset. Testing in both directions is crucial, as factors such as the volume and shape of the air cells can cause differences in their pressure conduction characteristics in different directions. This means that the pressure interference caused by the first air cell on the second air cell is not necessarily identical to the pressure interference caused by the second air cell on the first air cell. After obtaining a bidirectional passive pressure change test data set, advanced data analysis methods are applied to the data. For example, regression analysis techniques are used to establish regression models for the data in each direction. The proportional relationship between the passive air cell pressure change under different active air cell pressure changes is calculated, thereby obtaining the first-direction interference coefficient and the second-direction interference coefficient. Finally, the interference coefficients in these two directions are combined and weighted averaged to generate a passive pressure interference coefficient that comprehensively characterizes the pressure interference relationship of the neighboring air cell combination. This process is repeated for all neighboring air cell combinations, ultimately generating multiple passive pressure interference coefficients. These coefficients provide critical data support for accurately constructing the pressure interference relationship of neighboring air cells, enabling precise control of the pressure distribution of the smart mattress and improving the user's sleep experience.

[0026] The multiple passive pressure interference coefficients obtained are systematically integrated and analyzed. These coefficients are obtained through bidirectional passive pressure change testing and in-depth analysis of each combination of neighboring air cells. They contain information about the complex pressure interactions between adjacent air cells. A relational model is constructed using the air cells in the mattress as nodes, for example, a matrix-like neighboring air cell pressure interference matrix. In this matrix, rows and columns represent different air cells, and the matrix elements are the corresponding passive pressure interference coefficients. This matrix structure clearly and intuitively presents the degree and direction of each air cell's pressure influence on its neighboring air cells. For example, the element in row i and column j of the matrix represents the passive pressure interference coefficient of the jth air cell on the ith air cell. The magnitude of the value reflects the strength of the interference, while the sign indicates the direction of the pressure influence (a positive sign may indicate a pressure increase, a negative sign indicates a pressure decrease). This allows for precise prediction and compensation of pressure interference between air cells, thereby achieving precise optimization of the pressure distribution of the smart mattress, providing users with a more comfortable and ergonomic sleep support experience.

[0027] In one possible implementation, step S120 further includes: Step S121: extracting a first neighborhood air cell combination result from the plurality of neighborhood air cell combination results, and obtaining a first air cell and a second air cell in the first neighborhood air cell combination result.

[0028] Step S122: performing interference test analysis on the first air chamber and the second air chamber in the first direction and the second direction respectively, and generating a first direction interference coefficient and a second direction interference coefficient.

[0029] Step S123: generating a first passive pressure interference coefficient using the first direction interference coefficient and the second direction interference coefficient and adding the first passive pressure interference coefficient to the plurality of passive pressure interference coefficients.

[0030] Step S124: Similarly, traverse the plurality of neighboring air chamber combination results for analysis to generate the plurality of passive pressure interference coefficients.

[0031] Specifically, data extraction is first performed on the multiple previously generated neighborhood cell combination results. These combination results contain information on all potentially interfering cell pairs in the mattress. Using a specific indexing or identification mechanism, the first neighborhood cell combination result is precisely located and extracted. This combination result represents a specific combination of a pair of adjacent cells in the mattress. After successfully extracting the first neighborhood cell combination result, the internal structure of this combination is further analyzed. Using mattress internal structure mapping data or pre-defined cell identification rules, the two key cells within this combination result are accurately identified and retrieved, defined as the first and second cells. These two cells will serve as the core research objects in subsequent pressure interference testing and analysis. Their physical properties (such as volume, shape, elasticity, etc.) and relative positional relationships will significantly influence the pressure interference between them, laying the foundation for the subsequent accurate assessment of the pressure interference relationship between neighborhood cells. This is a key step in the entire smart mattress pressure optimization process.

[0032] A comprehensive and detailed bidirectional interference test analysis was conducted for the identified first and second chambers. First, in the first-direction test, the first chamber was set as the active interference source, and the second chamber as the passive response target. Using a high-precision and precisely controllable pressure generator, the pressure within the first chamber was gradually varied in small, preset increments, such as increasing the pressure by a certain value (e.g., 0.1 kPa) from the initial pressure. During the pressure change in the first chamber, the pressure change in the second chamber was monitored in real time using a highly sensitive pressure sensor. The sensor accurately captured extremely small pressure fluctuations, ensuring data accuracy. The corresponding pressure change data for each change in the first chamber's pressure was recorded, forming a series of data pairs to construct a first-direction passive pressure change test dataset. Based on this dataset, data analysis methods, such as linear regression analysis or curve fitting techniques, were used to determine the quantitative relationship between the pressure changes in the first and second chambers, and the proportionality coefficient between the two was calculated. This coefficient is known as the first-direction interference coefficient. This coefficient reflects the extent and characteristics of the interference of the first chamber on the second chamber's pressure in the first direction. After completing the test in the first direction, the test in the second direction is immediately carried out. At this time, the second air chamber is set as the active interference source, and the first air chamber is used as the passive response air chamber, and the above-mentioned process of pressure change, monitoring and data analysis is repeated. Similarly, the pressure of the second air chamber is gradually changed, and the passive change data of the pressure of the first air chamber is recorded to construct a second-direction pressure passive change test data set, and the second-direction interference coefficient is obtained by analyzing the data set. This coefficient reflects the degree and regularity of the pressure interference of the second air chamber on the first air chamber in the opposite direction. Through such a two-way interference test analysis, the pressure interference relationship between the first air chamber and the second air chamber in different directions can be fully and accurately grasped, providing key data support for the subsequent generation of a comprehensive passive pressure interference coefficient, thereby more accurately describing the complex pressure interactions between neighboring air chambers and providing a solid theoretical basis for the pressure distribution optimization of smart mattresses.

[0033] Based on the mattress's actual physical properties and patterns revealed by previous experimental data, an appropriate calculation method is selected to combine the interference coefficients in these two directions. For example, if experiments show that the interference in the two directions has similar impact, a simple averaging method is used: the interference coefficient in the first direction is added to the interference coefficient in the second direction and divided by 2 to obtain the first passive pressure interference coefficient. If the mattress structure or air cell characteristics make the interference in one direction more critical, a weighted averaging method can be used to assign corresponding weights to the interference coefficients in different directions. The weights are determined based on extensive experimental results and theoretical analysis to ensure that the actual pressure interference is accurately reflected. After the first passive pressure interference coefficient is obtained using the selected calculation method, it is added to the existing set of multiple passive pressure interference coefficients. This set provides the foundation for constructing a comprehensive and accurate data set for neighboring air cell pressure interference relationships. This addition process requires adherence to specific data management rules to ensure that the newly generated coefficient is correctly integrated with the other coefficients for subsequent unified analysis and processing. After the successful addition of the first passive pressure interference coefficient, the entire passive pressure interference coefficient set was further enriched and improved, more comprehensively covering the pressure interference information of different neighboring air chamber combinations in the mattress in different directions, and providing more sufficient data support for the precise construction of the neighboring air chamber pressure interference relationship in subsequent steps, which effectively promoted the smooth progress of the intelligent mattress pressure distribution optimization work, and ultimately achieved the goal of providing users with a comfortable and healthy sleep support environment.

[0034] A comprehensive and systematic traversal analysis is performed on all neighboring cell combinations. After completing the bidirectional interference test analysis for the first and second cells in the first neighboring cell combination, generating and adding the first passive pressure interference coefficient, attention is then shifted to the next neighboring cell combination. Following the same rigorous process, the two cells in this combination are extracted and again subjected to interference test analysis in the first and second directions. The corresponding directional interference coefficients are calculated, and the passive pressure interference coefficient for this combination is generated using the selected calculation method. This coefficient is then accurately added to the set of multiple passive pressure interference coefficients. This process is repeated continuously, with each neighboring cell combination result thoroughly analyzed, ensuring that no cell pair with a potential pressure interference relationship is missed. During this traversal, each cell may play the role of active or passive interference source in different combinations, fully accounting for the complex and diverse interactions between the various cells in the mattress. Through this comprehensive and meticulous traversal analysis, a complete set of multiple passive pressure interference coefficients is successfully generated. These coefficients fully characterize the pressure interference relationship between all neighboring air chambers in the mattress, laying a solid data foundation for the subsequent precise construction of the neighboring air chamber pressure interference relationship model. This enables the smart mattress to accurately optimize the pressure distribution in actual operation based on this accurate information, providing users with a highly personalized, comfortable and ergonomic sleeping experience.

[0035] In one possible implementation, step S122 further includes: Step S1221: performing a pressure passive change test with the first air chamber as an active interference item and the second air chamber as a passive interference item to generate a first direction passive interference test data set.

[0036] Step S1222: Analyze the pressure response interference relationship of the first air chamber to the second air chamber according to the first-direction passive interference test data set to generate the first-direction interference coefficient.

[0037] Step S1223: Perform a pressure passive change test with the second air chamber as an active interference item and the first air chamber as a passive interference item to generate a second direction passive interference test data set.

[0038] Step S1224: Analyze the pressure response interference relationship of the second air chamber to the first air chamber according to the second-direction passive interference test data set to generate the second-direction interference coefficient.

[0039] Specifically, the first air chamber is set as an active interference term, and the second air chamber as a passive interference term for passive pressure change testing. To this end, a high-precision pressure regulation device is installed, which can gradually change the pressure inside the first air chamber according to preset small pressure increments (such as 0.05kPa). At the same time, a highly sensitive pressure sensor is used to monitor the pressure changes inside the second air chamber in real time. The sensor records data at an extremely high sampling frequency (such as 100 times per second) to ensure that even the slightest pressure fluctuations can be captured. As the pressure in the first air chamber gradually increases from the initial value to the maximum value and then gradually decreases back to the initial value, the second air chamber pressure change data is continuously collected to generate a first-direction passive interference test data set. This data set contains multiple sets of different first air chamber pressure values ​​and the corresponding second air chamber pressure change values, comprehensively reflecting the pressure impact of the first air chamber on the second air chamber in the first direction.

[0040] Next, we conducted an in-depth analysis of the generated first-direction passive interference test data set. Using advanced data analysis algorithms, such as multivariate linear regression or polynomial fitting, we constructed a mathematical model using the pressure value of the first chamber as the independent variable and the pressure change of the second chamber as the dependent variable. By solving the model's parameters, we quantified the interference relationship between the pressure response of the first chamber and the second chamber, and calculated the first-direction interference coefficient. This coefficient intuitively reflects the average change in the second chamber's pressure for each unit change in the first chamber's pressure in the first direction and its trend, providing a key indicator for assessing the degree of interference between the two.

[0041] The roles were then reversed, with the second chamber acting as the active interference and the first chamber acting as the passive interference. Similarly, the pressure in the second chamber was varied using a high-precision pressure regulator, while the pressure changes in the first chamber were monitored by a highly sensitive pressure sensor. Data was collected to form a second-direction passive interference test dataset. This dataset records the response of the first chamber's pressure to changes in the second chamber's pressure, complementing the first-direction dataset to form a complete bidirectional pressure interference data set for neighboring chambers.

[0042] Finally, a similar analysis is performed based on the second-direction passive interference test data set. Using the same or similar data analysis method as step S1222, a mathematical model is constructed between the second chamber pressure and the first chamber pressure changes. The model parameters are solved to obtain the pressure response interference relationship of the second chamber on the first chamber, thereby generating the second-direction interference coefficient. This coefficient accurately describes the characteristics and degree of pressure interference of the second chamber on the first chamber in the opposite direction. Together with the first-direction interference coefficient, it comprehensively depicts the complex bidirectional pressure interference relationship between the first and second chambers. This provides indispensable data support for the subsequent generation of a comprehensive passive pressure interference coefficient and the construction of a pressure interference relationship model for the entire mattress neighborhood, which is of vital importance for achieving precise pressure distribution optimization in smart mattresses.

[0043] In one possible implementation, step S500 further includes: Step S510: Obtain basic physical information of the target user, wherein the basic physical information includes height, weight, and muscle distribution characteristics.

[0044] Step S520: Input the target reference spine alignment curve and the basic body information into an air pressure analysis network for analysis to obtain a target air pressure distribution map.

[0045] Step S530: Taking the target air pressure distribution map as a control target, performing interference compensation according to the interference relationship of the neighboring air chamber pressures, and generating the plurality of air chamber control parameters.

[0046] Specifically, basic physical information of the target user is obtained through various methods. Specialized measuring equipment, such as high-precision stadiometers, can be used to accurately obtain the user's height data, and scales can be used to accurately measure weight information. For muscle distribution characteristics, advanced bioelectrical impedance analysis or muscle imaging technologies can be used. These technologies can detect the content, density, and distribution of muscles in different parts of the human body through weak currents or specific imaging principles. This allows for a comprehensive understanding of the target user's basic physical information, providing a critical data foundation for subsequent precise pressure distribution optimization.

[0047] The target baseline spinal alignment curve and comprehensive body information are fed into a carefully constructed air pressure analysis network. This network is built based on an existing machine learning model. During its construction, a large amount of sample data from users of varying heights, weights, and muscle distributions in various sleeping positions was collected. This data includes the corresponding spinal alignment curves and the ideal air pressure distribution data collected by mattress pressure sensors. The machine learning model is trained using a supervised learning approach, using the spinal alignment curves and body information from the samples as feature vectors and the corresponding ideal air pressure distribution data as labels. Utilizing neural network architectures from deep learning algorithms, such as multi-layer perceptrons (MLPs) or convolutional neural networks (CNNs), the hidden layers of the neural network continuously adjust the connection weights between neurons by learning the complex mapping between features and labels from a large amount of sample data. During training, a backpropagation algorithm is used to propagate the error signal from the output layer to the input layer based on the error between the predicted air pressure distribution and the actual label. This updates the network parameters, leading to gradual model convergence and improved prediction accuracy. After a thorough training, the air pressure analysis network, when receiving the target baseline spinal alignment curve and the target user's basic body information, can perform complex calculations and reasoning based on the learned mapping relationship. Taking into account factors such as the weight distribution of various parts of the user's body, the physiological curvature of the spine, and the layout and characteristics of the mattress air chambers, it predicts the air pressure values ​​required for each area of ​​the mattress to achieve the best support effect, and then generates a target air pressure distribution map. This distribution map uses the smart mattress as a benchmark and accurately presents the air pressure required at different locations on the mattress surface. It provides a clear target orientation for subsequent precise control of multiple air chambers, ensuring that the mattress can provide the most appropriate pressure support based on the individual differences of the user, effectively promoting the user's sleep comfort and spinal health.

[0048] The target pressure profile serves as the core control objective. This profile clearly illustrates the ideal pressure required for optimal user support in each region of the mattress. Next, based on the neighboring cell pressure interference relationship established through rigorous testing and analysis, the interplay between cells is further considered. Because each cell in a mattress operates independently, changes in pressure in one cell can cause pressure disturbances in its neighboring cells. This disturbance can cause the actual pressure distribution to deviate from the target pressure profile. To achieve precise control, a disturbance compensation function is constructed based on the neighboring cell pressure interference relationship through complex mathematical modeling. This function quantifies the amount of compensation required for the target pressure of the current cell due to changes in the pressure of adjacent cells. Based on this function, an disturbance compensation analysis is performed in conjunction with the target pressure of each individual cell region in the target pressure profile. For each cell, the actual baseline control pressure that should be achieved is calculated after accounting for the interference of neighboring cells. Finally, based on these calculated baseline control pressures, cell control parameters are matched. Key control parameters within each cell, such as the output power of the air pump and the inflation or deflation volume of the airbag, are determined to generate multiple cell control parameters. These parameters will directly act on the air pump and airbags in the smart mattress to achieve precise control of the mattress pressure distribution, ensuring that the mattress can be as close as possible to the ideal state set by the target air pressure distribution map during actual operation, providing users with a highly ergonomic, comfortable and healthy sleeping support environment, effectively improving the user's sleep quality.

[0049] In one possible implementation, step S530 further includes: Step S531: determining target pressures of the plurality of independent air chamber regions according to the target air pressure distribution map.

[0050] Step S532: establishing an interference compensation function according to the neighboring air chamber pressure interference relationship.

[0051] Step S533: performing interference compensation analysis based on the interference compensation function and the target pressures of the multiple independent air chamber regions to generate multiple basic control pressures.

[0052] Step S534: matching control parameters of the air pumps and air bags in the multiple independent air chamber areas with the multiple basic control pressures, the multiple air chamber control parameters.

[0053] Specifically, the target air pressure distribution map is analyzed in depth. This map, based on the surface of the smart mattress, presents the ideal distribution of air pressure values ​​at various locations on the mattress in two or three dimensions, containing a wealth of pressure information. Pre-defined mattress area division rules are used to precisely divide the entire mattress surface into multiple sub-regions corresponding to independent air chambers. These division rules take into account the mattress's physical structure, air chamber layout, and the force distribution characteristics of the human body on the mattress, ensuring that each sub-region accurately corresponds to an independent air chamber, thereby establishing a mapping relationship between air pressure distribution and air chamber areas. Then, for each divided sub-region, the corresponding air pressure value is extracted from the target air pressure distribution map. This value represents the target pressure that the independent air chamber corresponding to that sub-region should ideally achieve. The extraction process involves data processing techniques, such as data interpolation algorithms, to process data points located at sub-region boundaries or special locations to ensure that the obtained target pressure values ​​are more accurate and representative. In this way, the target pressures of multiple independent air chamber areas can be accurately determined based on the target air pressure distribution map, providing a key data basis for subsequent interference compensation and control parameter generation, enabling the smart mattress to be precisely adjusted towards achieving the optimal pressure distribution, providing users with a comfortable and ergonomic sleep support experience.

[0054] This work was based on the pressure interference relationship between neighboring cells, as determined by in-depth previous research. Considering the complexity of the interactions between cells, mathematical modeling was used to quantify the pressure interference between cells into a functional form. Using the pressure change of neighboring cells as the independent variable and the pressure change of the current cell after the interference as the dependent variable, an interference compensation function was fitted using extensive experimental data. The coefficients of this function reflect the degree and direction of the pressure interference of different neighboring cells on the current cell, providing a theoretical basis for subsequent precise compensation.

[0055] Based on the determined target pressures for multiple independent air cell regions, these are used as baseline data for the interference compensation analysis process. Furthermore, the previously constructed interference compensation function is fully utilized. This function incorporates key information about the interference relationships between neighboring air cell pressures, such as the quantitative coefficients that determine the impact of neighboring air cell pressure changes on the current air cell pressure. A detailed interference compensation calculation begins for each independent air cell region. Sensors are used to obtain the current pressure state of the neighboring air cells within the region in real time, and the pressure change compared to the initial state is calculated. These pressure changes are then substituted into the interference compensation function. Combined with the target pressure for the region itself, the baseline control pressure required to achieve the target pressure is calculated, taking into account the interference from neighboring air cells. The calculation process comprehensively considers the impact of all neighboring air cells, which involves complex iterative calculations and data processing. For example, when a region has many neighboring air cells, the interference caused by the pressure change of each neighboring air cell is calculated sequentially and added to the total interference compensation. Through this rigorous calculation process, the corresponding baseline control pressure is ultimately generated for each independent air cell region. These basic control pressures will serve as an important basis for the subsequent matching of air pump and airbag control parameters, ensuring that the smart mattress can effectively cope with pressure interference between air chambers during actual operation, achieve precise optimization of pressure distribution, and provide users with a comfortable and ergonomic sleep support environment.

[0056] Work has begun on precisely matching control parameters for air pumps and airbags within multiple independent air chambers. First, the specific operating characteristics of the air pumps and airbags within each independent air chamber are thoroughly studied, including key performance parameters such as the pump's pressure-flow curve, pump efficiency curve, and the airbag's elastic modulus versus pressure curve. These parameters reflect the performance of the pump and airbag under different operating conditions and serve as the foundation for precise control. Next, mathematical models and algorithms are used to calculate control parameters based on the base control pressure and the operating characteristics of the pump and airbag. For the air pump, the base control pressure is combined with its pressure-flow curve to calculate control parameters such as the pump speed and operating time required to achieve that pressure. For example, a higher base control pressure requires a higher pump speed and longer operating time based on the pressure-flow curve. For the airbag, the base control pressure and its elastic modulus versus pressure curve are used to determine the amount of airbag inflation or deflation to generate corresponding control parameters, such as the opening time and degree of the solenoid valve. Throughout the control parameter matching process, the synergistic relationship between the air pump and airbags is fully considered, ensuring that their control parameters are mutually adapted to achieve precise regulation of pressure in each independent air chamber area. This ultimately generates multiple air chamber control parameters, which are directly input into the smart mattress's control system to direct the operation of the air pump and airbags, ensuring that the mattress provides comfortable and stable support according to the desired pressure distribution, effectively improving the user's sleep experience.

[0057] In one possible implementation, step S532 further includes: Step S5321: .

[0058] in, is the first Target pressure for each air chamber area; For the Basic control pressure of each air chamber area; For the Neighboring air cell regions of an air cell region Pressure changes; Neighboring air chamber area For the first Passive pressure interference coefficient of each air chamber area; For the The number of neighboring air cell areas of an air cell area.

[0059] Specifically, in the intelligent mattress pressure distribution optimization method with multiple air pumps working together, the interference compensation function is the key factor to achieve precise pressure control. For each air chamber area, the target pressure It is the pressure value that the air chamber should reach under ideal conditions based on the target air pressure distribution diagram, which is the goal pursued by the entire optimization process.

[0060] The basic control pressure It is the pressure that can be controlled by the air pump and airbag in the air chamber area without considering the interference of neighboring air chambers. It is an important parameter that needs to be determined through calculation. Indicates the Neighboring air cell regions of an air cell region The pressure change can be monitored in real time by the built-in pressure sensor of the mattress. It reflects the change of the current pressure of the neighboring air chamber relative to the initial state and is the key data for calculating the impact of interference. For the first Passive pressure interference coefficient of the air chamber area , is a quantitative index obtained through a large number of experiments and in-depth analysis of the pressure interference relationship between mattress air chambers. It accurately describes the pressure of neighboring air chambers. Air chamber The extent and nature of the pressure disturbance. It makes clear that The number of neighboring air chamber areas around each air chamber area. In practical applications, the neighborhood relationship of each air chamber area is first determined according to the design and layout of the mattress, so as to determine Then, the pressure change of each neighboring air chamber area is obtained through the pressure sensor , and combined with the existing passive pressure interference coefficient . Known target pressure , substitute these values ​​into the interference compensation function In the above example, we use the shift operation The basic control pressure can be calculated This calculation process is repeated for each independent air chamber area in the mattress. In this way, while taking into account the interference of neighboring air chambers, the basic control pressure of each air chamber area is determined, providing accurate control parameters for the subsequent coordinated operation of the air pump and airbags. This enables the smart mattress to achieve precise pressure distribution optimization based on the user's needs and physical condition, providing users with a comfortable and ergonomic sleeping support experience and effectively improving sleep quality.

[0061] In one possible implementation, step S200 further includes: Step S210: Visualizing the pressure distribution according to the pressure collection data to generate a mattress surface pressure distribution map, wherein the mattress surface pressure distribution map includes pressures at multiple positions.

[0062] Step S220: inputting the mattress surface pressure distribution map into a connected region detection model to perform connected region recognition and generate a contact surface posture recognition result.

[0063] Step S230: inputting the contact surface posture recognition result into a posture matching list for user posture matching to generate the user parsed posture, wherein the posture matching list includes multiple groups of contact surface posture samples and user posture samples having corresponding relationships.

[0064] Specifically, the acquired pressure data is systematically organized and analyzed. This pressure data is collected in real time by multiple high-precision pressure sensors distributed across the surface of the smart mattress. The data contains information about the pressure applied to each location on the mattress. These discrete data points are then converted into a continuous, intuitive graphical representation. Using the mattress surface as a two-dimensional coordinate system, each pressure sensor location corresponds to a point on the plane, and the pressure value collected at that point is represented using visual elements such as color, grayscale, or height. For example, higher pressure values ​​can be assigned a color closer to red and a higher brightness, or represented as a higher point in a three-dimensional visualization. In this way, the entire mattress surface is mapped and rendered point by point, ultimately generating a pressure distribution map. This map clearly and detailedly displays pressure conditions at multiple locations on the mattress surface, providing a clear and comprehensive understanding of the pressure distribution on the mattress. Information such as concentrated and dispersed pressure areas and the gradual pressure change can be intuitively presented, providing an intuitive, vivid, and accurate data visualization foundation for subsequent in-depth analysis of the user's posture on the mattress and further optimization of mattress pressure distribution.

[0065] The previously generated pressure distribution map on the mattress surface is fed into a specially designed connected region detection model. This model uses advanced image analysis techniques to comprehensively scan and analyze the pressure distribution map. The model identifies connected regions by identifying continuity and similarity in pressure values. In the mattress surface pressure distribution map, the areas where the body contacts the mattress form relatively concentrated regions with similar pressure values. These areas appear as continuous clusters of pixels on the image. The connected region detection model accurately locates these pixel clusters and labels them as independent connected regions. After identifying connected regions, the model further extracts and analyzes the features of each connected region, including geometric properties such as shape, size, position, perimeter, and centroid coordinates. These properties reflect the morphological characteristics of the contact area between the body and the mattress, such as the size of the contact area and the contour of the contact shape. Through comprehensive analysis of these connected region features, the contact surface posture recognition results are generated. The result describes in data form the posture information of the human body contact parts represented by each connected area in the mattress surface pressure distribution map. For example, a larger connected area may correspond to the human torso. Its shape and position information can help determine whether the human body is lying on its back, side or stomach, etc., providing key intermediate data for subsequent accurate matching of user postures, enabling the smart mattress to better understand the user's actual state on the mattress, and thus laying the foundation for personalized pressure distribution optimization.

[0066] The contact surface posture recognition results are obtained. These results contain posture feature information about the body contact area represented by each connected region in the mattress surface pressure distribution map. These results are then input into a pre-built posture matching list. The posture matching list is a rich data repository containing numerous sets of corresponding contact surface posture samples and user posture samples. These samples are carefully compiled through extensive experiments and data collection on users of various body types and sleeping postures. During the matching process, the input contact surface posture recognition results are compared in detail with each set of contact surface posture samples in the posture matching list, covering key features such as the shape, size, and positional relationship of the connected regions. The similarity between the two is calculated, and a distance metric algorithm is applied to identify the set of samples that most closely resembles the input results. Once the best matching set of samples is found, the user's current resolved posture can be accurately determined based on the pre-defined correspondence. This resolved posture accurately reflects the user's actual body posture on the mattress, such as whether they are lying flat on their back, lying on their left side, lying on their right side, lying prone, or in some transitional position between the two. This precise analysis of user posture provides a key basis for subsequent smart mattresses to optimize personalized pressure distribution according to the user's specific posture, ensuring that the mattress can adapt to and meet the user's sleep needs in real time, effectively improving the user's sleep comfort and experience.

[0067] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0068] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

[0069] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. An intelligent mattress pressure distribution optimization method with multiple air pumps working in coordination, characterized in that: include: Identify multiple independent air chamber areas in the smart mattress, perform pressure interference analysis on adjacent air chambers, and establish pressure interference relationships between adjacent air chambers; The surface pressure sensor network of the smart mattress is used to collect pressure data of the target user, and the contact surface is analyzed based on the pressure data to generate the user's analytical posture; Acquiring a preset spinal alignment curve database, wherein the preset spinal alignment curve database includes multiple postures of the target user and corresponding multiple reference spinal alignment curves; Inputting the user's analyzed posture into the preset spinal alignment curve database, comparing it with the multiple postures, and obtaining a target reference spinal alignment curve; Taking the target reference spinal alignment curve as the target, the pressure distribution of the multiple independent air chamber areas is optimized in combination with the pressure interference relationship of the neighboring air chambers, and multiple air chamber control parameters are generated to perform pressure distribution control.

2. The intelligent mattress pressure distribution optimization method with multiple air pumps working in coordination as claimed in claim 1, characterized in that: Identify multiple independent air chamber areas in the smart mattress, perform pressure interference analysis on adjacent air chambers, and establish pressure interference relationships between adjacent air chambers, including: Obtaining the distribution positions of the plurality of independent air chamber areas, and combining the air chambers at adjacent positions according to the distribution positions to generate a plurality of neighborhood air chamber combination results; Performing a bidirectional pressure passive change test and analysis on the adjacent air chambers based on the combination results of the plurality of adjacent air chambers to generate a plurality of passive pressure interference coefficients; The neighboring air cell pressure interference relationship is generated using the multiple passive pressure interference coefficients.

3. The method for optimizing pressure distribution of a smart mattress with multiple air pumps working in coordination as claimed in claim 2, characterized in that: The results of the combination of the plurality of adjacent air chambers are subjected to a bidirectional pressure passive change test and analysis of the adjacent air chambers to generate a plurality of passive pressure interference coefficients, including: Extracting a first neighborhood air cell combination result from the plurality of neighborhood air cell combination results, and obtaining a first air cell and a second air cell in the first neighborhood air cell combination result; Perform interference test analysis on the first air chamber and the second air chamber in a first direction and a second direction, respectively, to generate a first-direction interference coefficient and a second-direction interference coefficient; generating a first passive pressure interference coefficient using the first directional interference coefficient and the second directional interference coefficient and adding the first passive pressure interference coefficient to the plurality of passive pressure interference coefficients; Similarly, the plurality of adjacent air chamber combination results are traversed and analyzed to generate the plurality of passive pressure interference coefficients.

4. The method for optimizing pressure distribution of a smart mattress with multiple air pumps working in coordination as claimed in claim 3, characterized in that: For the first air chamber and the second air chamber, interference test analysis is performed in a first direction and a second direction respectively to generate a first direction interference coefficient and a second direction interference coefficient, including: Performing a pressure passive change test with the first air chamber as an active interference item and the second air chamber as a passive interference item to generate a first direction passive interference test data set; Analyze the pressure response interference relationship of the first air chamber to the second air chamber according to the first direction passive interference test data set to generate the first direction interference coefficient; Performing a pressure passive change test with the second air chamber as an active interference item and the first air chamber as a passive interference item to generate a second direction passive interference test data set; The second-direction interference coefficient is generated by analyzing the pressure response interference relationship of the second air chamber to the first air chamber according to the second-direction passive interference test data set.

5. The intelligent mattress pressure distribution optimization method with multiple air pumps working in coordination as claimed in claim 1, characterized in that: Taking the target reference spinal alignment curve as a target, the pressure distribution of the multiple independent air chamber areas is optimized in combination with the pressure interference relationship of the adjacent air chambers to generate multiple air chamber control parameters, including: Obtaining basic physical information of the target user, wherein the basic physical information includes height, weight, and muscle distribution characteristics; Inputting the target reference spine alignment curve and the basic body information into an air pressure analysis network for analysis to obtain a target air pressure distribution map; The target air pressure distribution diagram is used as a control target, interference compensation is performed according to the interference relationship of the neighboring air chamber pressures, and the plurality of air chamber control parameters are generated.

6. The intelligent mattress pressure distribution optimization method with multiple air pumps working in coordination as claimed in claim 5, characterized in that: Taking the target air pressure distribution map as the control target, performing interference compensation according to the interference relationship of the neighboring air chamber pressures to generate the multiple air chamber control parameters, including: determining target pressures of the plurality of independent air chamber regions according to the target air pressure distribution map; An interference compensation function is established based on the interference relationship of the neighboring air chamber pressures: performing an interference compensation analysis based on the interference compensation function and the target pressures of the plurality of independent gas chamber regions to generate a plurality of basic control pressures; The control parameters of the air pumps and air bags in the multiple independent air chamber areas are matched using the multiple basic control pressures, and the multiple air chamber control parameters.

7. The method for optimizing pressure distribution of a smart mattress with multiple air pumps working in coordination as claimed in claim 6, characterized in that: The interference compensation function is: ； in, is the first Target pressure for each air chamber area; For the Basic control pressure of each air chamber area; For the Neighboring air cell regions of an air cell region Pressure changes; Neighboring air chamber area For the first Passive pressure interference coefficient of each air chamber area; For the The number of neighboring air cell areas of an air cell area.

8. The method for optimizing pressure distribution of a smart mattress with multiple air pumps working in coordination as claimed in claim 1, characterized in that: The contact surface is analyzed based on the pressure data collected to generate the user's analytical posture, including: Visualizing pressure distribution based on the pressure collection data to generate a mattress surface pressure distribution map, wherein the mattress surface pressure distribution map includes pressures at multiple locations; Inputting the mattress surface pressure distribution map into a connected region detection model to perform connected region recognition and generate a contact surface posture recognition result; The contact surface posture recognition result is input into a posture matching list for user posture matching to generate the user parsed posture, wherein the posture matching list includes multiple groups of contact surface posture samples and user posture samples with corresponding relationships.

Images