Self-adaptive mattress adjusting system based on pressure induction

Abstract

The invention relates to mattress adjustment, in particular to a self-adaptive mattress adjustment system based on pressure sensing, which comprises a pressure sensing module for detecting pressure data, temperature data and acceleration data of each part of a human body; the pressure distribution modeling module is used for mapping the pressure data of each part of the human body into a three-dimensional pressure thermodynamic diagram of each part of the human body; the sleeping posture recognition module is used for recognizing a sleeping posture and a spine curve of the user according to the three-dimensional pressure thermodynamic diagram, judging a bedsore risk area in combination with the temperature data and the three-dimensional pressure thermodynamic diagram, and pre-judging whether the user has a turning-over intention in combination with the acceleration data and the three-dimensional pressure thermodynamic diagram; the adjusting instruction generation module is used for generating a corresponding adjusting instruction according to the sleeping posture of the user, the spine curve, the bedsore risk area and whether the user has a turning-over intention or not in combination with user preference historical data; according to the technical scheme provided by the invention, the defect that the sleep quality of a user is poor due to the fact that a mattress in the prior art cannot dynamically adapt to human body requirements can be effectively overcome.

Description

Technical Field

[0001] The present invention relates to mattress adjustment, and specifically to an adaptive mattress adjustment system based on pressure sensing. Background Art

[0002] Traditional mattresses have a fixed hardness and cannot adapt to different body shapes, sleeping postures, and health needs (such as local support required by patients with lumbar diseases). At the same time, uneven static pressure distribution causes discomfort during sleep and may lead to blood circulation disorders (such as bedsores) in the long term. Most existing smart mattresses rely on preset mode switching and lack the ability of real-time dynamic feedback adjustment. Summary of the Invention

[0003] (1) Technical Problems to be Solved

[0004] In view of the above-mentioned drawbacks of the prior art, the present invention provides an adaptive mattress adjustment system based on pressure sensing, which can effectively overcome the defect that the existing mattresses cannot dynamically adapt to human needs and cause poor sleep quality of users.

[0005] (2) Technical Solutions

[0006] To achieve the above object, the present invention is realized through the following technical solutions:

[0007] An adaptive mattress adjustment system based on pressure sensing, comprising a pressure sensing module, a pressure distribution modeling module, a sleeping posture recognition module, an adjustment instruction generation module, and an adjustment execution mechanism connected to a control unit;

[0008] The pressure sensing module detects pressure data, temperature data, and acceleration data of various parts of the human body;

[0009] The pressure distribution modeling module maps the pressure data of various parts of the human body into a three-dimensional pressure heat map of various parts of the human body;

[0010] The sleeping posture recognition module recognizes the user's sleeping posture and spinal curve according to the three-dimensional pressure heat map, combines the temperature data and the three-dimensional pressure heat map to judge the bedsore risk area, and combines the acceleration data and the three-dimensional pressure heat map to predict whether the user has the intention to turn over;

[0011] The adjustment instruction generation module generates corresponding adjustment instructions according to the user's sleeping posture, spinal curve, bedsore risk area, whether the user has the intention to turn over, and combines the user's preference historical data;

[0012] The adjustment execution mechanism controls the air pump connected to the airbag through double-loop PID control based on the pressure-support force mapping model and in combination with the adjustment instructions.

[0013] Preferably, the pressure sensing module detects the pressure data of various parts of the human body through a grid piezoresistive sensor covering the surface of the mattress. The grid piezoresistive sensor combines a hardware filtering circuit and a software sliding window mean algorithm to eliminate false triggers;

[0014] A temperature sensor and an inertial measurement unit IMU are integrated on the grid piezoresistive sensor;

[0015] Among them, the grid piezoresistive sensor communicates with the control unit through an SPI interface.

[0016] Preferably, the pressure distribution modeling module maps the pressure data of various parts of the human body into a three-dimensional pressure heat map of various parts of the human body, including:

[0017] Mapping the pressure data of various parts of the human body into a three-dimensional pressure heat map of the back, waist, and legs of the human body.

[0018] Preferably, the sleep posture recognition module identifies the user's sleep posture and spinal curve according to the three-dimensional pressure heat map, combines the temperature data and the three-dimensional pressure heat map to judge the bedsore risk area, and combines the acceleration data and the three-dimensional pressure heat map to predict whether the user has a turning intention, including:

[0019] Using a support vector machine (SVM) to identify the user's sleep posture according to the three-dimensional pressure heat map;

[0020] Identifying the spinal curve according to the three-dimensional pressure heat map;

[0021] When it is detected that the temperature exceeds the temperature threshold and the local pressure is greater than the pressure threshold for a certain period of time, it is judged that the user's blood circulation is blocked, and the bedsore risk area is determined;

[0022] When it is detected that the body movement acceleration is greater than the acceleration threshold and the pressure change rate is not less than the change rate threshold, it is judged that the user has a turning intention.

[0023] Preferably, the adjustment instruction generation module generates corresponding adjustment instructions according to the user's sleep posture, spinal curve, bedsore risk area, whether the user has a turning intention, and combines the user's preference historical data, including:

[0024] Combining the user's sleep posture and the pressure threshold to generate corresponding adjustment instructions;

[0025] Adjusting the airbag pressure in the lumbar region according to the spinal curve, maintaining the lumbar curvature error, and generating corresponding adjustment instructions;

[0026] Decompressing the airbag in the bedsore risk area according to the bedsore risk area, and generating corresponding adjustment instructions;

[0027] Pre-inflate the airbags in adjacent areas according to whether the user has the intention to turn over, reduce the support delay during position switching, and generate corresponding adjustment instructions.

[0028] Preferably, based on the pressure-support force mapping model, the adjustment actuator controls the air pump connected to the airbag through dual-loop PID control in combination with the adjustment instruction, including:

[0029] Based on the pressure-support force mapping model, control the air pump connected to the airbag through dual-loop PID control in combination with the adjustment instruction to accurately maintain the target air pressure value;

[0030] For the airbags in the non-active area, control the air pump connected to the airbag in an intermittent pressure-holding mode to reduce the overall power consumption;

[0031] Among them, the dual-loop PID control includes an inner loop for controlling the air pressure accuracy and an outer loop for adjusting the support hardness. The adjustment actuator communicates with the control unit through the CAN bus.

[0032] Preferably, the pressure-support force mapping model is expressed by the following formula:

[0033] F support =K1*P avg +K2*(dP / dt)+K3*T local ;

[0034] Among them, F support is the support force, P avg is the average pressure, dP / dt is the pressure change rate, T local is the local temperature, and K1, K2, and K3 are the weight coefficients of the average pressure, pressure change rate, and local temperature respectively. K1 = 0.6, K2 = 0.3, K3 = 0.1, and the weight coefficients K1, K2, and K3 are optimized through reinforcement learning Q-Learning.

[0035] Preferably, the airbag adopts a vertically stacked double-layer airbag. The inner layer airbag is inflated for support, and the outer layer airbag releases shear force through lateral deformation to reduce friction with the skin. Each airbag is independently connected to an air pump and a solenoid valve;

[0036] The air pump adopts an eight-zone independent air chamber design, and the flow rate can be adjusted.

[0037] Preferably, it further includes a sleep data statistics module, a mobile terminal, and an interactive touch screen connected to the control unit;

[0038] The sleep data statistics module judges the pressure stability according to the three-dimensional pressure heat map, calculates the sleep quality score in combination with the number of turns over, and sends it to the control unit together with the three-dimensional pressure heat Figure 1 map;

[0039] A mobile terminal that can visually display a three-dimensional pressure and heat map and a sleep quality score in real time, and can select scenario settings such as a medical mode, a sports recovery mode, a zero-pressure mode, and a reading mode for the control unit;

[0040] An interactive touch screen is provided on the side of the mattress and supports one-key emergency leveling and sensitivity settings;

[0041] Among them, the mobile terminal communicates with the control unit via Bluetooth 5.0 and uses AES-128 for data encryption transmission.

[0042] (III) Beneficial effects

[0043] Compared with the prior art, the pressure-sensing based adaptive mattress adjustment system provided by the present invention has the following beneficial effects:

[0044] 1) Multi-region pressure dynamic mapping and closed-loop control: Through high-density sensors and sub-region adjustment actuators, local hardness adjustment with millimeter-level response accuracy is achieved;

[0045] 2) Edge computing and lightweight algorithms: Data processing is completed within the embedded terminal to avoid cloud transmission delays and meet real-time requirements;

[0046] 3) User habit self-learning: Combining initial calibration (user input of weight and medical history) and historical data of user preferences formed during long-term use to optimize the adjustment strategy;

[0047] 4) Multi-modal health management expansion: Reserved interfaces are connected to heart rate / body movement sensors to implement derivative functions such as sleep apnea early warning;

[0048] Through the collaborative innovation of high-precision pressure sensors, intelligent algorithms, and sub-region adjustment execution, the present invention overcomes the defect that traditional mattresses cannot dynamically adapt to human needs, has broad application prospects in the fields of smart home, medical care, etc. At the same time, experiments show that the system can significantly improve the user's sleep quality and reduce the risk of long-term bedridden complications, combining technological advancement and market feasibility. Description of the drawings

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0050] Figure 1 It is a system schematic diagram of the present invention. Detailed implementation manners

[0051] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without any creative work belong to the scope of protection of the present invention.

[0052] The pressure-sensing adaptive mattress adjustment system, as Figure 1 shown, includes a pressure-sensing module, a pressure distribution modeling module, a sleeping posture recognition module, an adjustment instruction generation module, and an adjustment execution mechanism connected to a control unit;

[0053] The pressure-sensing module detects the pressure data, temperature data, and acceleration data of various parts of the human body;

[0054] The pressure distribution modeling module maps the pressure data of various parts of the human body into a three-dimensional pressure thermal map of various parts of the human body;

[0055] The sleeping posture recognition module recognizes the user's sleeping posture and spinal curve according to the three-dimensional pressure thermal map, combines the temperature data and the three-dimensional pressure thermal map to judge the bedsore risk area, and combines the acceleration data and the three-dimensional pressure thermal map to predict whether the user has the intention to turn over;

[0056] The adjustment instruction generation module generates corresponding adjustment instructions according to the user's sleeping posture, spinal curve, bedsore risk area, whether the user has the intention to turn over, and combines the user's preference historical data;

[0057] The adjustment execution mechanism controls the air pump connected to the airbag through dual-loop PID control based on the pressure-support force mapping model and in combination with the adjustment instruction.

[0058] ① The pressure-sensing module detects the pressure data of various parts of the human body through a grid-type piezoresistive sensor covering the mattress surface. The grid-type piezoresistive sensor combines a hardware filtering circuit (such as an RC low-pass filter) and a software sliding window mean algorithm to eliminate false triggers;

[0059] A temperature sensor (SHT35) and an inertial measurement unit IMU are integrated on the grid-type piezoresistive sensor;

[0060] Among them, the grid-type piezoresistive sensor communicates with the control unit through an SPI interface.

[0061] In the technical solution of this application, the grid piezoresistive sensor covers more than 90% of the mattress surface area with a 15*30 grid (a total of 450 independent detection points). The size of each grid is 5cm*5cm, the single-point detection accuracy is ±2kPa, and it supports a dynamic response at the 0.1s level.

[0062] ② The pressure distribution modeling module maps the pressure data of each part of the human body into a three-dimensional pressure heat map of each part of the human body, including:

[0063] Mapping the pressure data of each part of the human body into a three-dimensional pressure heat map of the human back, waist, and legs.

[0064] ③ The sleep posture recognition module identifies the user's sleep posture and spinal curve based on the three-dimensional pressure heat map, combines the temperature data and the three-dimensional pressure heat map to judge the bedsore risk area, and combines the acceleration data and the three-dimensional pressure heat map to predict whether the user has a turning intention, including:

[0065] Using the support vector machine SVM to identify the user's sleep posture (such as supine, side lying, etc.) based on the three-dimensional pressure heat map;

[0066] Identifying the spinal curve (vertebrae L1 - S1) based on the three-dimensional pressure heat map;

[0067] When it is detected that the temperature exceeds the temperature threshold (32°C), and the local pressure is greater than the pressure threshold (35 mmHg) for a certain period of time (5 min), it is judged that the user's blood circulation is not normal, and the bedsore risk area is determined;

[0068] When it is detected that the body movement acceleration is greater than the acceleration threshold (0.2g), and the pressure change rate is not less than the change rate threshold (1.5kPa / s), it is judged that the user has a turning intention.

[0069] ④ The adjustment instruction generation module generates corresponding adjustment instructions according to the user's sleep posture, spinal curve, bedsore risk area, whether the user has a turning intention, and combines the user's preference historical data, including:

[0070] Combining the user's sleep posture and the pressure threshold (such as when the waist pressure > 40kPa, triggering adjustment) to generate corresponding adjustment instructions;

[0071] Adjusting the airbag pressure in the lumbar region (12 - 18kPa) according to the spinal curve, maintaining the lumbar curvature error (error < 3°), and generating corresponding adjustment instructions;

[0072] Decompressing the airbag in the bedsore risk area according to the bedsore risk area, and generating corresponding adjustment instructions;

[0073] Pre-inflating the airbag in the adjacent area according to whether the user has a turning intention, reducing the support delay during position switching, and generating corresponding adjustment instructions.

[0074] ⑤Based on the pressure - support force mapping model, the adjustment actuator controls the air pump connected to the airbag through dual - loop PID control in combination with the adjustment instruction, including:

[0075] Based on the pressure - support force mapping model, the air pump connected to the airbag is controlled through dual - loop PID control in combination with the adjustment instruction to accurately maintain the target air pressure value (error < 2%);

[0076] For the airbags in the non - active area, an intermittent pressure - maintaining mode (that is, inflating for 1 s every 10 min to maintain a reference pressure of 5 kPa) is adopted to control the air pump connected to the airbag, so as to reduce the overall power consumption (the overall power consumption can be reduced by 40%);

[0077] Among them, the dual - loop PID control includes an inner loop for controlling the air pressure accuracy (±0.3 kPa) and an outer loop for adjusting the support hardness (error < 5%). The sampling period is 10 ms, and the adjustment actuator communicates with the control unit through the CAN bus (baud rate 1 Mbps, error rate < 0.01%).

[0078] Specifically, the pressure - support force mapping model (used to ensure that the support force is adaptively adjusted according to the pressure, pressure dynamic change and temperature) is expressed by the following formula:

[0079] F support =K1*P avg +K2*(dP / dt)+K3*T local ;

[0080] Among them, F support is the support force, P avg is the average pressure, dP / dt is the pressure change rate, T local is the local temperature, and K1, K2, K3 are the weight coefficients of the average pressure, pressure change rate, and local temperature respectively. K1 = 0.6, K2 = 0.3, K3 = 0.1. The weight coefficients K1, K2, K3 are optimized through reinforcement learning Q - Learning to reduce the number of times of user manual intervention.

[0081] In the technical solution of this application, the airbag adopts a vertically stacked double - layer airbag (80 mm * 80 mm). The inner - layer airbag is inflated for support, and the outer - layer airbag releases shear force through lateral deformation (compared with the traditional single - layer airbag, the shear stress is reduced by 60%), reducing the friction with the skin. Each airbag is independently connected to an air pump and a solenoid valve;

[0082] The air pump (working pressure is 0 - 30 kPa, response time < 50 ms, noise < 25 dB) adopts an eight - partition independent air chamber design, and the flow rate can be adjusted (the air pump flow rate is adjustable from 0 to 5 L / min).

[0083] In the technical solution of this application, it also includes a sleep data statistics module, a mobile terminal, and an interactive touch screen connected to the control unit;

[0084] The sleep data statistics module judges the pressure stability according to the three-dimensional pressure thermal map, calculates the sleep quality score in combination with the number of turns, and sends it to the control unit together with the three-dimensional pressure thermal Figure 1 map;

[0085] The mobile terminal visually displays the three-dimensional pressure thermal map and the sleep quality score in real time, and can select scenario settings such as medical mode (such as postoperative care pressure release), exercise recovery mode, zero-pressure mode, and reading mode for the control unit;

[0086] The interactive touch screen is set on the side of the mattress and supports one-key emergency leveling and sensitivity setting;

[0087] Among them, the mobile terminal communicates with the control unit through Bluetooth 5.0 and uses AES-128 for data encryption transmission.

[0088] In the technical solution of this application, the control unit uses the STM32H743VIT6 chip, runs the FreeRTOS system, supports multi-threaded processing (parallel execution of sensor data parsing, algorithm operation, adjustment instruction generation, and adjustment actuator control), and the delay < 5ms.

[0089] To better illustrate the beneficial effects of the technical solution of this application, Table 1 gives the relevant test indicators:

[0090] Table 1 Comparison of Scheme Effects

[0091] Index Technical solution of the present application Traditional solution Pressure detection accuracy ±2 kPa ±5 kPa (polyurethane sensor) Adjustment response time 0.5s ≥1.2s Decubitus prevention rate 98％ 82% (fluctuating air mattress) Number of night-time turnovers 4.3 times 8.7 times (fixed hardness mattress)

[0092] In terms of the technical solution of this application, from the three aspects of health benefits, user experience, and technical advantages:

[0093] Health benefits: The uniformity of pressure distribution is increased by 60% (compared with the experimental data of traditional mattresses); the local pressure overrun time of high-risk patients with pressure sores (such as long-term bedridden patients) is reduced by 85%;

[0094] User experience: The average improvement of the sleep quality score (PSQI) is 30%; the self-learning algorithm improves the adjustment accuracy from 70% to 95% within 7 days;

[0095] Technical advantages: The system power consumption < 15W, and the standby time exceeds 72h; it supports OTA upgrade and is compatible with third-party health management platforms.

[0096] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An adaptive mattress adjustment system based on pressure sensing, characterized in that: It includes a pressure sensing module, a pressure distribution modeling module, a sleeping posture recognition module, an adjustment instruction generation module, and an adjustment execution mechanism connected to a control unit; The pressure sensing module detects pressure data, temperature data, and acceleration data of various parts of the human body; The pressure distribution modeling module maps the pressure data of various parts of the human body into a three-dimensional pressure thermal map of various parts of the human body; The sleeping posture recognition module identifies the user's sleeping posture and spinal curve based on the three-dimensional pressure thermal map, combines the temperature data and the three-dimensional pressure thermal map to judge the pressure ulcer risk area, and combines the acceleration data and the three-dimensional pressure thermal map to predict whether the user has the intention to turn over; The adjustment instruction generation module generates corresponding adjustment instructions according to the user's sleeping posture, spinal curve, pressure ulcer risk area, whether the user has the intention to turn over, and combines the user's preference historical data; The adjustment execution mechanism controls the air pump connected to the airbag through double-loop PID control based on the pressure-support force mapping model and combines the adjustment instructions; 2. The pressure-sensing-based adaptive mattress adjustment system according to claim 1, characterized in that: The pressure sensing module detects the pressure data of various parts of the human body through a grid piezoresistive sensor covering the surface of the mattress, and the grid piezoresistive sensor combines a hardware filtering circuit and a software sliding window mean algorithm to eliminate false triggers; A temperature sensor and an inertial measurement unit IMU are integrated on the grid piezoresistive sensor; Among them, the grid piezoresistive sensor communicates with the control unit through an SPI interface.

3. The pressure-sensing-based adaptive mattress adjustment system according to claim 1, wherein: The pressure distribution modeling module maps the pressure data of various parts of the human body into a three-dimensional pressure thermal map of various parts of the human body, including: Mapping the pressure data of various parts of the human body into a three-dimensional pressure thermal map of the back, waist, and legs of the human body.

4. The pressure-sensing-based adaptive mattress adjustment system according to claim 1, characterized in that: The sleeping posture recognition module identifies the user's sleeping posture and spinal curve based on the three-dimensional pressure thermal map, combines the temperature data and the three-dimensional pressure thermal map to judge the pressure ulcer risk area, and combines the acceleration data and the three-dimensional pressure thermal map to predict whether the user has the intention to turn over, including: Using a support vector machine SVM to identify the user's sleeping posture based on the three-dimensional pressure thermal map; Identifying the spinal curve based on the three-dimensional pressure thermal map; When it is detected that the temperature exceeds the temperature threshold and the local pressure is greater than the pressure threshold for a certain period of time, it is judged that the user's blood circulation is not normal, and the pressure ulcer risk area is determined; When it is detected that the body movement acceleration is greater than the acceleration threshold and the pressure change rate is not less than the change rate threshold, it is judged that the user has the intention to turn over.

5. The pressure-sensing-based adaptive mattress adjustment system according to claim 1, wherein: The adjustment instruction generation module generates corresponding adjustment instructions according to the user's sleeping posture, spinal curve, pressure ulcer risk area, whether the user has the intention to turn over, and combines the user's preference historical data, including: Combining the user's sleeping posture and the pressure threshold to generate corresponding adjustment instructions; Adjusting the airbag pressure in the lumbar region according to the spinal curve, maintaining the lumbar curvature error, and generating corresponding adjustment instructions; Reducing the pressure of the airbag in the pressure ulcer risk area and generating corresponding adjustment instructions; Pre-inflating the airbag in the adjacent area according to whether the user has the intention to turn over, reducing the support delay during body position switching, and generating corresponding adjustment instructions.

6. The pressure-sensing based adaptive mattress adjustment system according to claim 1, wherein: The adjustment execution mechanism controls the air pump connected to the airbag through double-loop PID control based on the pressure-support force mapping model and combines the adjustment instructions, including: Based on the pressure-support force mapping model, the air pump connected to the airbag is controlled by dual-loop PID control in combination with adjustment instructions to accurately maintain the target air pressure value; For the airbags in the inactive area, an intermittent pressure-holding mode is used to control the air pump connected to the airbag to reduce the overall power consumption; Among them, the dual-loop PID control includes an inner loop for controlling air pressure accuracy and an outer loop for adjusting support hardness. The adjustment actuator communicates with the control unit through the CAN bus.

7. The pressure-sensing-based adaptive mattress adjustment system according to claim 6, wherein: The pressure-support force mapping model is expressed by the following formula: F support = K1*P avg + K2*(dP / dt) + K3*T local ; Among them, F support is the supporting force, P avg is the average pressure, dP / dt is the pressure change rate, T local is the local temperature, K1, K2, and K3 are the weight coefficients of the average pressure, the pressure change rate, and the local temperature respectively. K1 = 0.6, K2 = 0.3, K3 = 0.

1. The weight coefficients K1, K2, and K3 are optimized through the reinforcement learning Q-Learning.

8. The pressure-sensing-based adaptive mattress adjustment system according to claim 6, characterized in that: The airbag adopts a vertically stacked double-layer airbag. The inner layer airbag is inflated for support, and the outer layer airbag releases shear force through lateral deformation to reduce friction with the skin. Each airbag is independently connected to an air pump and a solenoid valve; The air pump adopts an eight-zone independent air chamber design, and the flow rate can be adjusted.

9. The pressure-sensing based adaptive mattress adjustment system according to claim 1, characterized in that: It also includes a sleep data statistics module, a mobile terminal, and an interactive touch screen connected to the control unit; The sleep data statistics module judges the pressure stability according to the three-dimensional pressure thermal map, calculates the sleep quality score in combination with the number of turns, and sends it to the control unit together with the three-dimensional pressure thermal map; The mobile terminal visually displays the three-dimensional pressure thermal map and the sleep quality score in real time, and can select scenario settings such as medical mode, exercise recovery mode, zero pressure mode, and reading mode for the control unit; The interactive touch screen is set on the side of the mattress and supports one-key emergency leveling and sensitivity setting; Among them, the mobile terminal communicates with the control unit through Bluetooth 5.0 and uses AES-128 for data encryption transmission.

Images