Control method, device, equipment, readable storage medium and program product of mattress

Abstract

The application relates to a control method and device of a mattress, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a reference back feature corresponding to a target object, and acquiring back support data detected by the target object; determining an interpolation weight based on a back pressure curve, and performing interpolation processing on the back support data according to the interpolation weight to obtain a pressure distribution map; determining a deviation degree between a back support feature of the pressure distribution map and the reference back feature; and controlling the supporting force of the mattress based on a control strategy corresponding to the deviation degree. The method can provide a supporting force suitable for the target object, so as to guarantee sleep quality and physical health.

Description

Technical Field

[0001] This application relates to the field of smart home technology, and in particular to a mattress control method, device, computer equipment, computer-readable storage medium, and computer program product. Background Technology

[0002] With the development of smart homes, smart mattresses have emerged. These adjustments are usually aimed at comfort, and existing smart mattresses offer several preset firmness modes or zoned support. However, most mattresses require users to manually select the mode.

[0003] Requiring each user to establish a target body condition template according to professional medical standards would be a cumbersome process and difficult to scale up. Therefore, users need to use mattress-specific control modes to dynamically adapt to their own physical condition. Summary of the Invention

[0004] Therefore, it is necessary to provide a mattress control method, device, computer equipment, computer-readable storage medium, and computer program product that can accurately detect the body state of the target object and provide appropriate support to ensure sleep quality and physical health.

[0005] In a first aspect, this application provides a method for controlling a mattress, the method comprising:

[0006] Obtain the reference back features corresponding to the target object, and obtain the back support data detected by the target object;

[0007] The interpolation weights are determined based on the back pressure curve, and the back support data is interpolated according to the interpolation weights to obtain the pressure distribution map.

[0008] Determine the deviation between the back support feature of the pressure distribution map and the reference back feature;

[0009] The support force of the mattress is controlled based on the control strategy corresponding to the deviation.

[0010] In one embodiment, obtaining the reference back features corresponding to the target object includes:

[0011] After adjusting the target object's support data to undetermined support data according to the adjustment instructions within the first time period, and obtaining the historical evaluation results corresponding to the undetermined support data, the target support data and the undetermined support data are selected based on the historical evaluation results; a reference back feature is determined based on the selected support data; and / or,

[0012] In the second duration, historical support data of the target object at multiple times are determined, and the indicator trends of the target object at multiple times are determined; based on each indicator trend, historical support data at each time point are selected, and reference back features are determined based on the selected support data; the first duration is shorter than the second duration; and / or,

[0013] Based on the support data with a negative correlation to seasonal temperature and the support data of the enhanced region corresponding to the season, the reference back features corresponding to the target object are determined.

[0014] In one embodiment, determining the historical supporting data of the target object at multiple times includes:

[0015] The integrity and signal quality of the first historical support data of the target object at the first moment are detected, and the data quality positively correlated with the integrity and signal quality is determined.

[0016] Determine the parameter stability of multiple first historical support data at the first moment;

[0017] Obtain the historical evaluation results corresponding to the first historical support data;

[0018] The system confidence level at the second moment is obtained by weighted fusion of the data quality, the parameter stability, and the historical evaluation results corresponding to the first historical support data.

[0019] The first historical support data is adjusted according to the adjustment speed of the negative correlation of the system confidence level to obtain the second historical support data at the second time point; the first time point is one or more times before the second time point.

[0020] In one embodiment, determining the interpolation weights based on the back pressure curve includes:

[0021] Determine the back surface of the target object;

[0022] Add the back support data to the back surface to obtain a back pressure surface containing support data for each point.

[0023] The interpolation weights of the support data at each point are determined based on the curvature of the region in the back pressure surface.

[0024] In one embodiment, determining the interpolation weight of the support data at each point based on the regional curvature of the support data at each point in the back pressure surface includes:

[0025] Based on the curvature of the region in the back pressure surface according to the support data at each point, determine the interpolation kernel width that is negatively correlated with the curvature of the region;

[0026] The width of the interpolation kernel is positively correlated with the range of the support data interpolation at each point.

[0027] In one embodiment, determining the deviation between the back support feature of the pressure distribution map and the reference back feature includes:

[0028] Extract back support features in each dimension from the pressure distribution map;

[0029] Based on the feature differences between the back support features of the same dimension and the reference back features, the weighted deviation of each dimension is determined;

[0030] The deviation to be weighted is weighted according to the weight of each dimension to obtain the weighted deviation; wherein each dimension includes the key segment pressure value of the spinal axis, the symmetry index of both sides of the body and the non-core weight, and the weight of the key segment pressure value and the symmetry index is higher than that of the non-core weight.

[0031] The control strategy based on the deviation, which controls the support force of the mattress, includes:

[0032] Based on the weighted deviation, the magnitude of the negative correlation with the weighted deviation is determined, and the support force of the mattress is controlled.

[0033] Secondly, this application also provides a mattress control device, comprising:

[0034] The acquisition module is used to acquire the reference back features corresponding to the target object and to acquire the back support data detected by the target object;

[0035] The distribution map determination module is used to determine the interpolation weights based on the back pressure curve, and to perform interpolation processing on the back support data according to the interpolation weights to obtain the pressure distribution map;

[0036] A deviation determination module is used to determine the deviation between the back support feature of the pressure distribution map and the reference back feature;

[0037] The control module is used to control the support force of the mattress based on the control strategy corresponding to the deviation.

[0038] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the mattress control method in any of the above embodiments.

[0039] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the mattress control method in any of the above embodiments.

[0040] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the mattress control method in any of the above embodiments.

[0041] The aforementioned mattress control method, device, computer equipment, computer-readable storage medium, and computer program product acquire reference back features corresponding to the target object. These reference back features reflect the unique spinal structure and suitable physiological curvature of each target object. The method also acquires back support data detected by the target object to determine its current spinal state. Furthermore, it determines interpolation weights based on the back pressure curve and interpolates the back support data according to these weights to obtain a pressure distribution map. This ensures that the back support data is interpolated not only based on the distance of points but also on the pressure between the spine and the mattress, making the pressure distribution map more ergonomic and biomechanical, providing appropriate support for each part of the body. The method further determines the deviation between the back support features of the pressure distribution map and the reference back features. Based on the control strategy corresponding to this deviation, the mattress can have an adaptive support adjustment mechanism. Thus, by accurately analyzing the back features, it can determine whether the target object deviates from the optimal posture and provide support adapted to the target object, ensuring sleep quality and physical health. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is an application environment diagram of the mattress control method in one embodiment;

[0044] Figure 2 This is a flowchart illustrating a mattress control method in one embodiment;

[0045] Figure 3 This is a flowchart illustrating a reference back feature in one embodiment;

[0046] Figure 4 This is a flowchart illustrating the process of determining interpolation weights in one embodiment;

[0047] Figure 5 This is a structural block diagram of the control device for a mattress in one embodiment;

[0048] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0050] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various types of data, but these data are not limited by these terms. These terms are only used to distinguish between the first type of data and the second type of data. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the solutions, or any combination of multiple solutions.

[0051] The mattress control method provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, mattress 102 communicates with a local gateway or server 104 via a network. A data storage system can store the data that the local gateway or server 104 needs to process. The data storage system can be integrated into the local gateway or server 104, or it can be located in the cloud or on other network servers.

[0052] The mattress 102 includes a high-resolution pressure sensor array and a support execution system. Exemplarily, the sensor array is uniformly distributed in a dot matrix pattern, with a density of 150 sensing points per square meter, and the sampling frequency is adjustable within the range of 1-20Hz. The support execution system comprises sixteen independently controlled support units, employing an air pump-airbag structure. Each unit has a pressure adjustment range of 0-100 kPa and a response time of less than 1 second.

[0053] The local gateway or server 104 comprises three core engines: a template management engine responsible for storing, matching, and optimizing user templates; a heatmap analysis algorithm that uses an improved bicubic interpolation algorithm to generate a pressure distribution map and extract key feature values; and a control decision engine that optimizes the control parameters of the support unit based on a differential evolution algorithm. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0054] The interaction and presentation layer is implemented through the Terminal 106 application, providing a real-time visual interface displaying stress distribution maps, template comparisons, and health scores. It collects user comfort scores and subjective feelings, and offers long-term data tracking and personalized suggestion generation capabilities. Terminal 106 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, head-mounted displays, etc. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc.

[0055] In one exemplary embodiment, such as Figure 2 As shown, a mattress control method is provided, which is applied to Figure 1 Taking the local gateway or server 104 as an example, the explanation includes the following steps 202 to 208. Wherein:

[0056] Step 202: Obtain the reference back features corresponding to the target object, and obtain the back support data obtained from the detection of the target object.

[0057] The target audience is at least one user of the mattress. The target audience can be users of multiple age groups or body types, each with their own corresponding reference back features to accommodate their body shape or needs. For example, a 12-year-old user is one target audience, and a 20-year-old user is another.

[0058] The reference back features are the body features corresponding to the target object. The reference back features represent the set of back features that best fit the target object, providing a reference for back support. Optionally, the reference back features can be a set of feature data obtained by filling a standard template.

[0059] Back support data refers to the detection results of the support status of the target object's back. Back support data can be obtained based on mattress sensors, or it can be obtained by filtering sensor data. For example, a median filter can be used to remove impulse noise from the sensor data, a Gaussian filter can be used to smooth the spatial fluctuations of the noise-removed data, and the smoothed spatial fluctuation data can be normalized based on body weight to eliminate the influence of individual weight differences on the absolute pressure value.

[0060] Optionally, if the target subject continues to use the mattress, back support data can be determined based on data detected by sensors on the mattress. For example, whenever the target subject uses the mattress continuously for 1 hour, the sensor data within that hour is processed using methods such as mean filtering, and the result of the mean filtering is converted into back support data according to the location of each sensor. Continuous use refers to the period of intermittent use being less than a threshold. For example, if the target subject uses the mattress continuously for 10 minutes, and then leaves the mattress for 1 minute before continuing to use it, this is still considered continuous use.

[0061] In some embodiments, a standard template of the target object can be filled or modified based on user settings and historical back support data of the target object to obtain a personalized data set of the target object; based on the personalized data set, a reference back feature of the target object can be determined.

[0062] Step 204: Determine the interpolation weights based on the back pressure curve, and perform interpolation processing on the back support data according to the interpolation weights to obtain the pressure distribution map.

[0063] The back pressure curve is a back model adapted to the target object, representing the shape of the target object's contact with the mattress and reflecting the pressure distribution generated by the mattress supporting the target object. Optionally, the back pressure curve can be multiple curves, which not only represent the pressure distribution of the human spine, but also reflect the pressure distribution on both sides of the spine.

[0064] Interpolation weights are set according to the back pressure curve; these weights can adaptively adjust as the back pressure curve distribution changes. Back support data includes data points distributed at different locations, and the interpolation weights represent the degree of pressure influence on each data point. These weights determine the contribution of each data point to the pressure distribution map. Unlike traditional linear interpolation, the interpolation weights determined by the back pressure curve ensure that back support data is interpolated not only according to the distance of the points but also according to the pressure between the spine and the mattress. This makes the pressure distribution map more ergonomic and biomechanical, avoiding distorted pressure diffusion at the spinal groove and providing appropriate support for each part of the body.

[0065] A pressure distribution map shows the pressure distribution of a mattress at different locations. Optionally, it can not only represent the spatial distribution of back pressure, but also determine the changes in back pressure distribution at different times. A pressure distribution map can be a pressure graph composed of multiple pressure lines, or a pressure heat map. In a pressure heat map, each coordinate has a color, and the intensity of the color at each coordinate indicates the pressure value of the mattress at that coordinate. For example, a warmer color (closer to red) indicates a higher pressure value at that coordinate, while a cooler color (closer to blue) indicates a lower pressure value at that coordinate.

[0066] In some embodiments, the back support data includes data points for supporting the user, each data point representing the support force at that point; the interpolation weight of each data point can be determined based on the curvature of the back pressure curve; and bilinear or multilinear interpolation is performed on each data point according to the interpolation weight to obtain a pressure distribution map.

[0067] Step 206: Determine the deviation between the back support features of the pressure distribution map and the reference back features.

[0068] Back support features are the characteristic dimensions of how a mattress supports the back, extracted from pressure distribution maps. Because heatmap features represent the spatial pressure distribution and can reflect regional pressure and its changes, the extracted features accurately describe the spinal support status, more closely matching the actual back support situation, and facilitating feature comparison and difference analysis.

[0069] Deviation is a quantitative result of the current support state deviating from the ideal support state. The initial deviation is obtained by quantifying the difference between the current spinal support state and the ideal state through the vector difference between the back support features and the reference back features. Optionally, the initial deviation can be weighted based on weighting coefficients of different features, and the control strategy corresponding to the weighted deviation can be used to control the mattress support force.

[0070] Step 208: Control the support force of the mattress based on the control strategy corresponding to the deviation.

[0071] The control strategy refers to the level of support a mattress provides at various points. Different deviations correspond to different control strategies, which can result in variations in the support provided by the mattress in at least one area.

[0072] Support force includes at least the support force at each point. The corresponding support force can be set by adjusting the inflation level of the airbags at each point. Single-zone refers to adjusting a single independent support unit (e.g., adjusting only the airbag directly below the corresponding lumbar vertebra). This adjustment is typically used to correct single-point, multi-point, or localized support problems. Multi-zone refers to adjusting multiple interrelated support units simultaneously or sequentially (e.g., adjusting airbags in adjacent lumbar and pelvic regions simultaneously). This adjustment is used to address overall posture or support problems involving a large area or requiring coordination of multiple zones for correction.

[0073] Optionally, the mattress support control process may also involve three types of suggestions: real-time posture correction providing real-time guidance for detected poor posture; long-term health suggestions offering lifestyle improvement recommendations based on trend analysis; and product usage suggestions optimizing mattress usage based on usage pattern data. These suggestions are generated by a trained AI language model, and the frequency and timing of suggestion pushes are adaptively adjusted based on user acceptance and feedback.

[0074] In one embodiment, a closed-loop logic of "perception-decision-execution-verification" is employed. The system continuously collects stress data and calculates the current deviation. When a threshold is exceeded, an adjustment plan is generated and executed. After adjustment, the system waits for stabilization and then remeasures the next deviation. If the adjustment is ineffective, it is rolled back and lessons learned from the failure are accumulated. This mechanism ensures that each adjustment is based on actual effectiveness verification.

[0075] In one embodiment, the terminal displays the support results, scoring results, and trend results. Optionally, a real-time comparison display can be used, in which a heat map of the reference feature and a real-time pressure distribution map of the back support feature are displayed side by side, and a semi-transparent overlay technique is used to visually display the areas of difference. The color coding system uses a gradient color scheme from blue to red to represent pressure from low to high.

[0076] In some embodiments, a deviation-based health scoring system exists, in which a daily spinal health index is calculated based on the deviation, with a score ranging from 0 to 100. The scoring algorithm considers three factors: the average deviation at night, the maximum deviation, and the duration of high deviation.

[0077] In some embodiments, the terminal displays trend charts. By showing changes in long-term health scores, supporting data adjustment history, and correlation analysis of sleep quality, it helps users understand the impact of lifestyle habits on spinal health.

[0078] In the aforementioned mattress control method, reference back features corresponding to the target object are obtained. These features reflect the unique spinal structure and suitable physiological curvature of each target object. Back support data detected by the target object is also acquired to determine its current spinal state. Furthermore, interpolation weights are determined based on the back pressure curve, and the back support data is interpolated according to these weights to obtain a pressure distribution map. This ensures that the back support data is interpolated not only based on the distance of points but also on the pressure between the spine and the mattress, making the pressure distribution map more ergonomic and biomechanical, providing appropriate support for each part of the body. The deviation between the back support features of the pressure distribution map and the reference back features is then determined. Based on the control strategy corresponding to this deviation, the mattress can have an adaptive support adjustment mechanism. Thus, by accurately analyzing the back features, it is possible to determine whether the target object deviates from the optimal posture, providing support tailored to the target object to ensure sleep quality and physical health.

[0079] In some embodiments, such as Figure 3 As shown, obtaining the reference back features corresponding to the target object includes at least one of steps 302 to 306. These three steps can exist independently or be used in combination. Wherein:

[0080] Step 302: After adjusting the target object's support data to be adjusted to undetermined support data according to the adjustment instructions within the first time period, and obtaining the historical evaluation results corresponding to the undetermined support data, select the support data to be adjusted and the undetermined support data according to the historical evaluation results; determine the reference back features based on the selected support data.

[0081] The first duration is shorter than the second duration. The adjustment instruction is the mattress pressure adjustment data entered by the user, indicating whether the user expects the mattress to increase or decrease support in a certain area. The support data to be adjusted is the target object's support data before adjustment; this data can be the result of an adjustment or selection prior to the support data to be determined, or it can be a template entered or selected by the target object. The support data to be determined is the result obtained by adjusting according to the adjustment instruction. The support data to be adjusted differs from the support data to be determined; however, because the specific support data converted from the user's subjective feelings may not match the user's actual mattress usage, the corresponding evaluation results allow for more detailed control of the adjustment strategy.

[0082] Historical evaluation results are derived from: determining reference characteristics at historical moments using undetermined support data; determining the historical deviation between the back support characteristics of the historical pressure distribution map and the reference back characteristics; and controlling the mattress support force based on the control strategy corresponding to this historical deviation, and then basing the results on user input or the results obtained from various indicator tests. Optionally, historical evaluation results include positive and negative evaluations, where a positive evaluation indicates that the adjusted undetermined support data is better, and a negative evaluation indicates that the unadjusted undetermined support data is better.

[0083] In one example, short-term learning updates periodically based on adjustments and effect scores from 7 days of data. After each adjustment, effect data is collected for 3 nights. The adjustment is retained only if the effect is positive for 2 consecutive nights; otherwise, it is rolled back to the state before adjustment.

[0084] Step 304: In the second duration, determine the historical support data of the target object at multiple times and determine the indicator trend of the target object at multiple times; select the historical support data at each time according to the indicator trend, and determine the reference back features based on the selected support data; the first duration is shorter than the second duration.

[0085] Historical support data comprises data from multiple points in time. The historical support data at each point in time determines the reference back features for the corresponding historical time period, thereby enabling the execution of control processes at that historical moment. Optionally, sufficient high-quality data can be ensured to support the analysis, based on the completeness and accuracy of data within a second time period. Furthermore, representative historical support data is selected according to the analytical needs to more accurately detect the trends of key indicators of the target object. Optionally, historical support data includes support data to be adjusted and support data to be determined.

[0086] The indicator trend includes historical supporting data values ​​for the assessment indicators. Assessment indicators are the dimensions that need optimization. Optionally, assessment indicators include, but are not limited to, comfort scores, sleep quality indices, morning pain reports, and long-term postural stability.

[0087] For example, the comfort score is a subjective rating (1-10) submitted by the user through the app. Sleep quality index: This comprehensively considers sleep continuity and the frequency of unnecessary turning over detected during the night, for example, 0.4 * (longest continuous sleep period / total time in bed) + 0.6 * (1 - max(0, (actual turning frequency - baseline turning frequency) / baseline turning frequency)). Morning pain reports can be user-labeled pain locations (cervical spine, thoracic spine, lumbar spine, sacrum, pain level 1-10). Long-term postural stability can be the average of the standard deviations of pressure distribution every half hour during the night.

[0088] Optionally, the weighting coefficients of the indicators can be dynamically adjusted according to user type: for example, healthy young people prioritize improving their sleep experience, with comfort at 0.4, sleep quality at 0.35, pain report at 0.1, and postural stability at 0.15; sub-healthy middle-aged people prioritize comfort at 0.3, sleep quality at 0.3, pain report at 0.25, and postural stability at 0.15; those with a history of spinal diseases prioritize symptom relief, with comfort at 0.2, sleep quality at 0.25, pain report at 0.4, and postural stability at 0.15; and the elderly prioritize comfort at 0.25, sleep quality at 0.3, pain report at 0.3, and postural stability at 0.15.

[0089] In some embodiments, the changes in multiple indicators at adjacent time points are weighted and fused according to their respective weights to obtain historical indicators. Based on the changes in these historical indicators, historical indicators at their peak values ​​are selected from the historical support data at each time point. The reference backside feature at the current time point is determined based on the historical indicators at their peak values, and subsequent steps are performed accordingly. Thus, by using the changes in these historical indicators, the reference backside feature corresponding to the optimal indicator is determined more accurately.

[0090] In one example, long-term adaptation extracts all nighttime data from the past 90 days, calculates trends for various metrics, and identifies the support configurations for optimal performance.

[0091] Step 306: Based on the support data of the negative correlation between seasonal temperature and the support data of the enhanced region corresponding to the season, determine the reference back features corresponding to the target object.

[0092] Seasonal temperature indicates the temperature range for that season, such as the temperature range for summer or winter. Enhanced zone support data indicates support data for specific zones to provide additional support. For example, in warmer seasons, a smaller overall support data allows for a softer overall mattress area and enhanced support in ventilated zones, based on the determined reference back characteristics. Conversely, in colder seasons, a smaller overall support data allows for a firmer overall mattress area and enhanced support in specific zones to combat muscle tension caused by cold.

[0093] Optionally, both the support data negatively correlated with seasonal temperature and the support data for the enhanced regions corresponding to the season can include historical support data, which includes support data to be adjusted and support data to be determined. This creates an adjustment method across three time scales to more precisely determine the reference back features. Optionally, the enhanced region support data refers to areas related to factors such as the woman's menstrual cycle, daytime activity intensity, water intake, temperature, and humidity as reflected in wearable device data.

[0094] In an exemplary embodiment, if an adjustment instruction exists, the support data selected from historical support data is used as the pending support data; if no adjustment instruction exists, the reference back feature is determined based on the support data selected from historical support data.

[0095] In an exemplary embodiment, if there is seasonal variation, the historical support data is adjusted based on the support data negatively correlated with seasonal temperature and the support data of the enhanced region corresponding to the season to obtain the adjusted historical support data; the support data selected from the adjusted historical support data is used as the undetermined support data; if there is no seasonal variation, the reference back feature is determined based on the support data selected from the unadjusted historical support data.

[0096] In this embodiment, based on the evaluation results and feedback within the first time period, users can directly adjust the support data according to their own feelings, allowing for precise adjustments based on user experience or short-term indicators. Based on historical evaluation results within the second time period, the system adapts to changes in physical condition over the long term, dynamically identifying the optimal support configuration. Seasonal adjustments consider the impact of environmental factors on support needs. Thus, three selectable time scales are formed, making the support data more tailored to user needs and dynamically adjusting support capacity.

[0097] In some embodiments, determining historical support data for a target object at multiple time points includes: detecting the integrity and signal quality of first historical support data for the target object at a first time point, and determining data quality positively correlated with integrity and signal quality; determining the parameter stability of first historical support data at multiple first time points; obtaining historical evaluation results corresponding to the first historical support data; performing weighted fusion on data quality, parameter stability, and historical evaluation results corresponding to the first historical support data to obtain system confidence at a second time point; adjusting the first historical support data according to the adjustment rate negatively correlated with the system confidence level to obtain second historical support data at the second time point; the first time point is one or more time points prior to the second time point.

[0098] The first moment and the second moment are historical moments when the target object used the mattress. The first moment precedes the second moment, and the first historical support data used in the first moment is used to obtain the second historical support data used in the second moment, so that the support data used at each moment is different.

[0099] The first historical support data is the support data at the first moment, used to determine the reference characteristics at the first moment, and is used in conjunction with the back support data at the first moment to control the support force of the mattress at the first moment.

[0100] The second historical support data is the support data at a second moment, used to determine the reference characteristics at that second moment, and is used in conjunction with the back support data at the second moment to control the mattress support force at that moment. The second historical support data is the result of adjustments based on the first historical support data.

[0101] Integrity refers to whether there are any missing contents in the first historical support data obtained by the server, in order to determine the validity of the collected data. Optionally, the integrity of the collected data can be determined, for example, based on the ratio of effective collection time to total bedtime; alternatively, the mattress processor can treat the sensor data as a polynomial and perform division using a preset generator polynomial to obtain a fixed-length checksum, which can be a remainder, and append it to the original data before sending it. After receiving the data, the server recalculates the checksum using the same method and then compares it with the received checksum; if they are inconsistent, it indicates that the data may have been corrupted during transmission, thus determining the integrity of the data transmission.

[0102] Signal quality refers to the transmission quality supporting data transmission to the server. Signal quality is related to factors such as noise and signal transmission method. High signal quality means that the proportion of effective information in the data is high, and features can be extracted more reliably. Optionally, the signal quality can be determined by the ratio of the proportion of abnormal data points to the noise power ratio, and the reciprocal of this ratio can be used to determine the signal quality.

[0103] Data quality is a weighted fusion of the integrity of the initial historical supporting data and signal quality; data quality is positively correlated with both integrity and signal quality. Optionally, the weight of integrity is greater than that of signal quality to more accurately determine data quality. When data quality is poor, the supporting data needs to be adjusted quickly to adapt it to the target audience, allowing users to stay in bed longer and enhancing data integrity.

[0104] Parameter stability represents the trend of change between different first historical support data, reflecting the degree of deviation of each data value from the mean or preset value. It can be parameter volatility. When the data volatility is low, it needs to be adjusted quickly so that the support data can be adapted to the target object as soon as possible to ensure stability.

[0105] The historical evaluation results corresponding to the first historical supporting data include at least one average evaluation result at the first moment. The historical evaluation results corresponding to the first historical supporting data can be positive or negative, and both positive and negative evaluations can include multiple levels.

[0106] System confidence indicates the degree of match between the mattress and the target object. System confidence is positively correlated with data quality, parameter stability, and historical evaluation results corresponding to the first historical support data. Since these three data points are all dynamically changing, they are used to control the speed of template adaptive adjustment.

[0107] In some embodiments, the template evolution rate employs an adaptive algorithm, learning rapidly when the system confidence is low and fine-tuning for optimization when the confidence is high. For example, rapid learning (weight 0.25) is used during the initial period (first month) of mattress use by the target object and at extremely low confidence (<0.3), experimenting with different configurations; major adjustments require user confirmation. At medium confidence (0.3-0.7), automatic fine-tuning is possible with a moderate learning speed (weight 0.1), and at high confidence (>0.7), stability is maintained with minor adjustments (weight 0.03).

[0108] In one example, the system confidence score is calculated by combining the following factors: data quality (weighted at 0.4), parameter stability (weighted at 0.3), and historical evaluation results (weighted at 0.3). The system confidence score is obtained by weighting these three types of data.

[0109] The expression for data quality is: DQ = 0.6 × DI + 0.4 × Qs; where DQ represents data quality, DI represents data integrity, and Qs represents signal quality. The ratio of effective acquisition time to total bedtime represents integrity; data quality is the ratio of the proportion of anomalous data points to the noise power ratio, and the reciprocal of this ratio determines signal quality.

[0110] The expression for parameter stability is: Fp = 1 / (1 + σ); where Fp is the parameter volatility and σ is the standard deviation of the template parameter over the past 7 days.

[0111] Historical evaluation results are based on the adjustment made according to the view. If the user clearly approves the adjustment (comfort score greater than 7), the score is 0.3; if the feedback is average or there is no feedback, the score is 0.2; if the user is dissatisfied (comfort score less than 5) or manually adjusts, the score is -0.4.

[0112] In this embodiment, the signal quality, parameter stability, and historical evaluation results corresponding to the first historical support data are comprehensively considered to more meticulously evaluate the adjustment speed, thereby ensuring processing speed. The template evolution rate adopts an adaptive algorithm, which learns quickly when the system confidence is low and fine-tunes and optimizes when the confidence is high. The stress distribution characteristics corresponding to high-scoring nights are integrated into the reference features of the personal template, gradually improving the matching degree between the support strategy and personal needs.

[0113] In some embodiments, such as Figure 4 As shown, the interpolation weights are determined based on the back pressure curve, including:

[0114] Step 402: Determine the back surface of the target object.

[0115] The target individual's body shape indicates the distribution of weight and pressure points on the mattress, which may affect whether the spine can maintain its natural alignment. For example, heavier individuals or those with pronounced curvatures require a more supportive mattress, which provides a firmer feel and helps maintain the spine's natural curvature to prevent lumbar collapse. Lighter individuals or those with a straighter build may need a slightly softer mattress to distribute pressure evenly and avoid discomfort and pressure concentration on prominent areas such as the shoulders and hips due to an overly firm surface.

[0116] A back surface is a surface composed of the intersection of multiple back pressure curves, used to represent the back shape of a target object. The back surface can be determined based on a standardized pressure distribution template, and it can represent the distribution of pressure in different regions.

[0117] Step 404: Add the back support data to the back surface to obtain a back pressure surface containing support data for each point.

[0118] A back pressure surface is a back surface containing support data at various points. It represents the pressure distribution points when a target object uses the mattress, characterizing the user's current pressure on the mattress. For example, the back surface includes back curve 1, back curve 2, and back curve 3; back curve 1 and back curve 2 have three intersection points. Sensor data at these intersection points is added to form a back pressure surface containing support data at each point.

[0119] In some embodiments, adding back support data to the back surface to obtain a back pressure surface containing support data for each point includes: adding the pressure data of each support point to the back surface according to the coordinates of the sensors to which each support point belongs in the back support data, thereby obtaining a back pressure surface containing support data for each point.

[0120] Step 406: Determine the interpolation weight of each point's support data based on the regional curvature of each point's support data in the back pressure surface.

[0121] Regional curvature refers to the curvature of the region to which the supporting data belongs. When using regional curvature, the weight coefficients of each supporting data can be dynamically adjusted based on factors such as the user's weight, to more accurately control the interpolation weights. Regional curvature represents the natural degree and direction of the spine's curvature in a specific region, such as the curvature of the cervical, vertebral, or lumbar vertebrae.

[0122] In some embodiments, the interpolation weight of each point support data is determined based on the regional curvature of each point support data in the back pressure surface, the surface distance of each point support data along the back, and the human body region to which each point support data belongs.

[0123] In this embodiment, the corresponding interpolation weight is determined based on the curvature of the area where the support point is located, so that the interpolation weight is more in line with the spinal structure of the target object, accurately representing the pressure distribution, thereby avoiding distorted pressure diffusion in places such as the spinal groove.

[0124] In some embodiments, determining the interpolation weight of each point support data based on the regional curvature of each point support data in the back pressure surface includes: determining the interpolation kernel width that is negatively correlated with the regional curvature based on the regional curvature of each point support data in the back pressure surface; wherein the interpolation kernel width is positively correlated with the range of interpolation of each point support data.

[0125] The interpolation kernel width represents the range of support data points during interpolation. A smaller curvature in the region with a wider interpolation kernel width requires fewer support data points around that region for interpolation; conversely, a larger curvature requires more support data points around that region for interpolation. The resulting heatmap provides more detailed local information in each region, allowing for more accurate extraction of the corresponding back support features. The interpolation kernel can be either linear or bicubic linear interpolation.

[0126] In one example, at the high-curvature spinal groove, the interpolation kernel width is automatically narrowed to limit pressure diffusion to both sides and significantly reduce the weight of sensor points on the opposite side of the spine, thereby avoiding distortion caused by pressure crossing the midline of the spine. This mimics the real biomechanical principle—pressure is strongly transmitted longitudinally along the spine, while transmission across the spine laterally is very weak.

[0127] In one example, based on the interpolation rules adjusted according to the three-dimensional shape and physiological curvature of the human back, instead of simply calculating the weights based on planar distance, priority is given to the actual distance along the back curvature, reducing the lateral diffusion of pressure in high curvature areas (such as the spinal groove), and respecting anatomical boundaries (such as the midline of the spine and bony prominences), so that the pressure distribution map can more accurately represent the actual pressure on the target object.

[0128] In this embodiment, the spine structure of the target object is reflected by the interpolation kernel width of the negative correlation of regional curvature, so as to more accurately determine the range of support data at each point during interpolation, and enable the pressure distribution map to more accurately determine the pressure distribution of each local area.

[0129] In some embodiments, determining the deviation between the back support features of the pressure distribution map and the reference back features includes: extracting back support features of each dimension from the pressure distribution map; determining the weighted deviation of each dimension based on the feature differences between the back support features of the same dimension and the reference back features; and weighting the weighted deviation according to the weight of each dimension to obtain the weighted deviation. Each dimension includes the key segment pressure value of the spinal axis, the symmetry index of both sides of the body, and non-core weights, with the key segment pressure value and the symmetry index having higher weights than the non-core weights.

[0130] Correspondingly, based on the control strategy corresponding to the deviation, the support force of the mattress is controlled, including: based on the weighted deviation, determining the magnitude of the negative correlation with the weighted deviation, and controlling the support force of the mattress.

[0131] The feature difference is the weighted deviation in each dimension. By measuring the vector difference between the back support features and the reference back features in the same dimension, the feature difference between the current spinal support state and the ideal state can be quantified, thus obtaining the weighted deviation. The weighted deviation is the initial deviation calculated for each dimension, which can represent the difference between the ideal state and the actual state.

[0132] The weights of each dimension are set according to their correlation with stress. The weights differ across dimensions, which helps determine which dimension's deviation has a stronger effect. The weighted deviation is the sum of the deviations from multiple dimensions.

[0133] The critical segment pressure value represents the pressure value at one or more locations along the spine. The symmetry index represents the pressure difference between the left and right sides of the spine. Non-core weights include the coordinates of the regional pressure center of gravity and the high-pressure area.

[0134] In some embodiments, regional pressure includes pressure values ​​at key segments along the spinal axis, and regional pressure includes a left-right symmetry index. The pressure values ​​at key segments along the spinal axis include pressure distribution along at least one segment of the spinal axis, thus constituting a dimension with higher weighting. For example, the average pressure of the cervical segment, upper thoracic segment, lower thoracic segment, lumbar segment, and sacral segment; the left-right symmetry index may include the left-right pressure difference of the cervical region, thoracic region, and lumbar region.

[0135] In some embodiments, the dimensions of non-core weights include the coordinates of the regional pressure centroid and the high-pressure region, and the regional pressure change values ​​include the pressure gradient change. The coordinates of the regional pressure centroid include, but are not limited to, the X and Y coordinates of the pressure centers of the cervical spine, thoracic spine, and lumbar spine regions; while the high-pressure region includes, but is not limited to, the overall high-pressure area percentage and the area of ​​the local maximum high-pressure area. The pressure gradient change can be the overall pressure gradient change rate.

[0136] In some embodiments, the average pressure of a region is calculated by partitioning the area to obtain the pressure distribution along the spinal axis. The weighted average coordinates of all pressure points within a partition can be used to obtain the coordinates of the region's pressure centroid. The average pressure difference between the left and right sides of each region can be calculated separately, and the left-right symmetry index can be determined based on the average pressure difference between each region. High-pressure regions can be determined based on the percentage of pixel areas where the statistical pressure exceeds a specific threshold and the pixel area of ​​the largest continuous high-pressure region. The pressure gradient change rate can be determined based on the average rate of pressure change between adjacent pixels on the heatmap.

[0137] In some embodiments, when the weighted deviation is in the first interval, the support force can be monitored and adjusted; when the weighted deviation is in the second interval, the support force can be fine-tuned with a small adjustment range; when the weighted deviation is in the third interval, the support force can be adjusted in multiple regions; and when the weighted deviation is in the fourth interval, a significant adjustment can be performed. The first, second, third, and fourth intervals are weighted deviation intervals from small to large.

[0138] In one exemplary embodiment, four intervention levels are determined based on the four intervals of the weighted deviation: Level 1 intervention maintains the current support and only monitors when the weighted deviation is less than 0.1; Level 2 intervention initiates a fine-tuning mode when the weighted deviation is between 0.1 and 0.3, with the adjustment range in a single area not exceeding 5%; Level 3 intervention performs moderate adjustment when the weighted deviation is between 0.3 and 0.6, with coordinated changes in multiple areas; and Level 4 intervention performs significant adjustment when the weighted deviation exceeds 0.6, combined with user wake-up prompts.

[0139] In one example, the difference between the real-time feature vector and the personal template feature vector is calculated to obtain the weighted deviation. Based on the weight of each dimension, the weighted deviation is weighted and comprehensively scored to quantify the gap between the current spinal support state and the ideal state, thus obtaining the weighted deviation.

[0140] In some embodiments, both the back support feature and the reference back feature include individual data points. The back support feature includes the current value of the support force generated by each data point, while the reference back feature includes the template value of the support force generated by each data point. The system calculates the absolute value of the difference between the current value and the template value for each feature to obtain the single-point weighted deviation for that feature. Then, a pre-set weighting coefficient is assigned to each weighted deviation. These weighting coefficients follow clear clinical principles: core features directly related to spinal health, such as pressure values ​​at key segments of the spinal axis and the symmetry index of the left and right sides of the body, are given higher weights. This means that when the support force in these key areas needs adjustment, they contribute more to the final deviation, thus driving the system to prioritize and correct these support issues crucial to spinal physiological alignment. Finally, the weighted deviations of all features are summed to obtain the final value used to determine the weighted deviation for the control strategy. This value comprehensively and objectively reflects the degree of conformity between the current mattress support state and the individual's ideal health state, serving as the fundamental basis for triggering subsequent intelligent adjustments.

[0141] In this embodiment, the pressure values ​​of key segments of the spinal axis and the deviation of the symmetry index of both sides of the body have high dimensional weights. When there are support data deviations in these key dimensional areas, the weighted deviation changes more significantly, allowing the mattress to prioritize and correct these support issues that are crucial for the physiological alignment of the spine. Moreover, by integrating multiple dimensions through weighted deviation, the mattress objectively reflects the deviation between its current support status and the individual's ideal health status, enabling fine-tuning.

[0142] In one exemplary embodiment, a control method for a personalized smart mattress that balances accuracy and practicality is provided. Specific objectives include: providing users with a progressive support solution from general to personalized, forming corresponding reference back features through three levels: preset templates, user calibration, and continuous learning, thereby achieving a balance between accuracy and convenience, and continuously optimizing the support strategy so that the system can adapt to long-term changes in the user's physical condition. Based on this, a real-time correlation analysis mechanism between pressure heatmaps and spinal support status is established. Through closed-loop control, the mattress support units are automatically adjusted to maintain or approach a healthy curvature of the spine in a lying position. The system detects pressure distribution deviations in real time and automatically corrects them. Through visual feedback and intelligent suggestions, users are guided to develop healthy sleeping posture habits. The system provides real-time pressure distribution display, health scores, and personalized suggestions to enhance user health awareness.

[0143] Regarding the preset templates and user calibration, in some embodiments, based on more than 5,000 cases of spinal curvature and corresponding pressure distribution data measured by professional physicians, the K-means clustering algorithm is used to divide the user group into sixteen typical categories, a standardized pressure distribution template is established for each category, and then the initial standard template is determined by template matching.

[0144] The standard template approach includes basic parameters input by the user upon first use, such as age, gender, height, weight, and self-perceived health issues. After calculating the Body Mass Index (BMI), the system calculates the matching degree with each template category based on multi-dimensional features, and selects the most suitable preset template using a weighted scoring algorithm. The matching degree calculation comprehensively considers the similarity of demographic characteristics, body shape parameters, and health concerns. For example, a 30-year-old male, 180cm tall, and weighing 75kg will be matched with the average health stress distribution pattern of young men of similar body type in the database; a 50-year-old female, 160cm tall, weighing 65kg, who claims to have "back problems," will be matched with a template of a middle-aged woman of similar body type with stronger lumbar support.

[0145] The user feature values ​​are standardized, and a weight coefficient is assigned to each feature. For example, age is 0.25, gender is 0.25, BMI is 0.3, and the health complaint matching degree is 0.2. The similarity between the user feature vector and the feature vector of each preset template is calculated, and the template with the highest total similarity score is selected as the initial matching result. The classification feature table of the preset templates is shown in Table 1.

[0146] Table 1

[0147]

[0148] Based on the preset template, data can be further updated using a three-step calibration method. First, static baseline data is collected by having the user lie supine and remain still for 30 seconds, during which the system collects baseline pressure data. Next, dynamic response testing is performed, with the system sequentially adjusting each support unit to test the response to changes in pressure distribution. Finally, subjective preferences are recorded, allowing the user to experience three fine-tuning modes and select their preferred option. The calibration algorithm employs a weighted fusion strategy, where the preset template accounts for 70%, the static baseline for 20%, and the subjective preference adjustment for 10%, resulting in the calibrated template. This weighting distribution respects medical statistical principles while also considering individual measured data and subjective feelings.

[0149] Subjective preference involves choosing your favorite from three preset fine-tuning modes: Uniform Support: Pressure distribution across all areas tends to be more even, reducing pressure gradients. Differentiated Support: Strengthens support in key areas (e.g., lumbar region +5%, cervical spine +3%), allowing other areas to be relatively soft. Moderate Support: Slightly softens the overall support based on uniform support (e.g., all areas -2%).

[0150] The static baseline pressure data serves as a reference and calibration standard. The dynamic response adjustment of each unit is based on the change in this baseline. The subjective preference is to fine-tune the pressure based on the baseline pressure to ensure compliance with the healthy support principle. Optionally, the change in pressure distribution before and after adjustment yields a response sensitivity matrix. The sensitivity matrix is ​​the ratio between the pressure change and the adjustment amount, reflecting the degree to which the adjustment of each support unit alters the pressure distribution. The minimum adjustment amount required to produce a perceptible effect is calculated, providing control gain parameters for the closed-loop control algorithm, thereby forming the corresponding adjustment range.

[0151] This embodiment includes a mechanism for continuous learning and real-time correlation analysis, comprising:

[0152] If an adjustment instruction is obtained within the first time period, the target object's support data to be adjusted is adjusted to undetermined support data according to the adjustment instruction within the first time period. After obtaining the historical evaluation results corresponding to the undetermined support data, the support data to be adjusted and the undetermined support data are selected according to the historical evaluation results, and the reference back features are determined based on the selected support data.

[0153] If no adjustment instruction is obtained within the first time period, then in the second time period, the parameter stability of the first historical support data at multiple first moments is determined; the historical evaluation results corresponding to the first historical support data are obtained; the data quality, parameter stability, and the historical evaluation results corresponding to the first historical support data are weighted and fused to obtain the system confidence at the second moment; the first historical support data is adjusted according to the adjustment speed negatively correlated with the system confidence to obtain the second historical support data at the second moment; the first moment is one or more moments before the second moment; the indicator trend of the target object at multiple first moments and the second moment is determined; the historical support data at each moment is selected according to the trend of each indicator, and the reference back features are determined according to the selected support data; the first time period is shorter than the second time period.

[0154] Based on the support data with negative correlation to seasonal temperature and the support data of the enhanced regions corresponding to the season, the selected support data is adjusted to obtain the reference back features corresponding to the target object.

[0155] Obtain the back support data obtained from the target object detection; determine the back surface of the target object; add the back support data to the back surface to obtain the back pressure surface containing the support data of each point; determine the interpolation kernel width that is negatively correlated with the curvature of the region based on the regional curvature of the support data of each point in the back pressure surface; wherein, the interpolation kernel width is positively correlated with the range of interpolation of the support data of each point.

[0156] The back support data is interpolated according to the interpolation weights to obtain a pressure distribution map. Back support features of each dimension are extracted from the pressure distribution map. Based on the feature differences between the back support features of the same dimension and the reference back features, the deviation to be weighted for each dimension is determined. The deviation to be weighted is weighted according to the weight of each dimension to obtain the weighted deviation. Each dimension includes the pressure value of the key segment of the spinal axis, the symmetry index of both sides of the body, and the non-core weight. The pressure value of the key segment and the symmetry index have higher weights than the non-core weight.

[0157] Based on the weighted deviation, the magnitude of the negative correlation with the weighted deviation is determined to control the support force of the mattress.

[0158] In this embodiment, precise support based on individual anatomical characteristics is achieved, with pressure distribution matching the personal template by over 90%, demonstrating high long-term learning stability. Clinical trials show that users' deep sleep time increased by an average of 18 minutes, nighttime turning frequency decreased by 30%, and sleep continuity was significantly improved. This embodiment overcomes reliance on specialized medical equipment, making precise spinal health management more accessible. Through early intervention and daily management, it reduces the frequency of medical visits for spinal-related problems. Users can better understand their own spinal health status through system feedback, guiding them to engage in proactive health behaviors such as posture correction and targeted exercises, forming a virtuous cycle of health management.

[0159] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0160] Based on the same inventive concept, this application also provides a mattress control device for implementing the mattress control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more mattress control device embodiments provided below can be found in the limitations of the mattress control method described above, and will not be repeated here.

[0161] In one exemplary embodiment, such as Figure 5 As shown, a mattress control device is provided, comprising:

[0162] The acquisition module 502 is used to acquire the reference back features corresponding to the target object and acquire the back support data detected by the target object;

[0163] The distribution map determination module 504 is used to determine the interpolation weight based on the back pressure curve, and to perform interpolation processing on the back support data according to the interpolation weight to obtain a pressure distribution map;

[0164] The deviation determination module 506 is used to determine the deviation between the back support feature of the pressure distribution map and the reference back feature;

[0165] The control module 508 is used to control the support force of the mattress based on the control strategy corresponding to the deviation.

[0166] In one embodiment, the acquisition module 502 is configured to:

[0167] After adjusting the target object's support data to undetermined support data according to the adjustment instructions within the first time period, and obtaining the historical evaluation results corresponding to the undetermined support data, the target support data and the undetermined support data are selected based on the historical evaluation results; a reference back feature is determined based on the selected support data; and / or,

[0168] In the second duration, historical support data of the target object at multiple times are determined, and the indicator trends of the target object at multiple times are determined; based on each indicator trend, historical support data at each time point are selected, and reference back features are determined based on the selected support data; the first duration is shorter than the second duration; and / or,

[0169] Based on the support data with a negative correlation to seasonal temperature and the support data of the enhanced region corresponding to the season, the reference back features corresponding to the target object are determined.

[0170] In one embodiment, the acquisition module 502 is configured to:

[0171] The integrity and signal quality of the first historical support data of the target object at the first moment are detected, and the data quality positively correlated with the integrity and signal quality is determined.

[0172] Determine the parameter stability of multiple first historical support data at the first moment;

[0173] Obtain the historical evaluation results corresponding to the first historical support data;

[0174] The system confidence level at the second moment is obtained by weighted fusion of the data quality, the parameter stability, and the historical evaluation results corresponding to the first historical support data.

[0175] The first historical support data is adjusted according to the adjustment speed of the negative correlation of the system confidence level to obtain the second historical support data at the second time point; the first time point is one or more times before the second time point.

[0176] In one embodiment, the distribution map determination module 504 is configured to:

[0177] Determine the back surface of the target object;

[0178] Add the back support data to the back surface to obtain a back pressure surface containing support data for each point.

[0179] The interpolation weights of the support data at each point are determined based on the curvature of the region in the back pressure surface.

[0180] In one embodiment, the distribution map determination module 504 is configured to:

[0181] Based on the curvature of the region in the back pressure surface according to the support data at each point, determine the interpolation kernel width that is negatively correlated with the curvature of the region;

[0182] The width of the interpolation kernel is positively correlated with the range of the support data interpolation at each point.

[0183] In one embodiment, the deviation determination module 506 is configured to:

[0184] Extract back support features in each dimension from the pressure distribution map;

[0185] Based on the feature differences between the back support features of the same dimension and the reference back features, the weighted deviation of each dimension is determined;

[0186] The deviation to be weighted is weighted according to the weight of each dimension to obtain the weighted deviation; wherein each dimension includes the key segment pressure value of the spinal axis, the symmetry index of both sides of the body and the non-core weight, and the weight of the key segment pressure value and the symmetry index is higher than that of the non-core weight.

[0187] The control module 508 is used for:

[0188] Based on the weighted deviation, the magnitude of the negative correlation with the weighted deviation is determined, and the support force of the mattress is controlled.

[0189] The various modules in the control device of the aforementioned mattress can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0190] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with an external mattress and terminal via a network connection. When the computer program is executed by the processor, it implements a mattress control method.

[0191] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0192] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0193] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0194] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0195] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0196] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0197] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0198] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for controlling a mattress, characterized in that, The method includes: Obtain the reference back features corresponding to the target object, and obtain the back support data detected by the target object; The interpolation weights are determined based on the back pressure curve, and the back support data is interpolated according to the interpolation weights to obtain the pressure distribution map. Determine the deviation between the back support feature of the pressure distribution map and the reference back feature; The support force of the mattress is controlled based on the control strategy corresponding to the deviation.

2. The method according to claim 1, characterized in that, The step of obtaining the reference back features corresponding to the target object includes: After adjusting the target object's support data to undetermined support data according to the adjustment instructions within the first time period, and obtaining the historical evaluation results corresponding to the undetermined support data, the target support data and the undetermined support data are selected based on the historical evaluation results; a reference back feature is determined based on the selected support data; and / or, In the second duration, historical support data of the target object at multiple times are determined, and the indicator trends of the target object at multiple times are determined; based on each indicator trend, historical support data at each time point are selected, and reference back features are determined based on the selected support data; the first duration is shorter than the second duration; and / or, Based on the support data with a negative correlation to seasonal temperature and the support data of the enhanced region corresponding to the season, the reference back features corresponding to the target object are determined.

3. The method according to claim 2, characterized in that, The determination of the historical supporting data of the target object at multiple times includes: The integrity and signal quality of the first historical support data of the target object at the first moment are detected, and the data quality positively correlated with the integrity and signal quality is determined. Determine the parameter stability of multiple first historical support data at the first moment; Obtain the historical evaluation results corresponding to the first historical support data; The system confidence level at the second moment is obtained by weighted fusion of the data quality, the parameter stability, and the historical evaluation results corresponding to the first historical support data. The first historical support data is adjusted according to the adjustment speed of the negative correlation of the system confidence level to obtain the second historical support data at the second time point; the first time point is one or more times before the second time point.

4. The method according to claim 1, characterized in that, The determination of interpolation weights based on the back pressure curve includes: Determine the back surface of the target object; Add the back support data to the back surface to obtain a back pressure surface containing support data for each point. The interpolation weights of the support data at each point are determined based on the curvature of the region in the back pressure surface.

5. The method according to claim 4, characterized in that, The step of determining the interpolation weight of the support data at each point based on the regional curvature of the support data at each point in the back pressure surface includes: Based on the curvature of the region in the back pressure surface according to the support data at each point, determine the interpolation kernel width that is negatively correlated with the curvature of the region; The width of the interpolation kernel is positively correlated with the range of the support data interpolation at each point.

6. The method according to claim 1, characterized in that, Determining the deviation between the back support feature of the pressure distribution map and the reference back feature includes: Extract back support features in each dimension from the pressure distribution map; Based on the feature differences between the back support features of the same dimension and the reference back features, the weighted deviation of each dimension is determined; The deviation to be weighted is weighted according to the weight of each dimension to obtain the weighted deviation; wherein each dimension includes the key segment pressure value of the spinal axis, the symmetry index of both sides of the body and the non-core weight, and the weight of the key segment pressure value and the symmetry index is higher than that of the non-core weight. The control strategy based on the deviation, which controls the support force of the mattress, includes: Based on the weighted deviation, the magnitude of the negative correlation with the weighted deviation is determined, and the support force of the mattress is controlled.

7. A control device for a mattress, characterized in that, The device includes: The acquisition module is used to acquire the reference back features corresponding to the target object and to acquire the back support data detected by the target object; The distribution map determination module is used to determine the interpolation weights based on the back pressure curve, and to perform interpolation processing on the back support data according to the interpolation weights to obtain the pressure distribution map; A deviation determination module is used to determine the deviation between the back support feature of the pressure distribution map and the reference back feature; The control module is used to control the support force of the mattress based on the control strategy corresponding to the deviation.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images