Self-adaptive intelligent sleep-aiding sleeping bag and sleep-aiding control method

Abstract

The invention relates to intelligent sleep health equipment, in particular to a self-adaptive intelligent sleep-aiding sleeping bag and a sleep-aiding control method. The sleeping bag comprises a sleeping bag body, and a physiological sensing layer, a multi-element intervention execution layer and a central processing system are integrated on the sleeping bag body. The physiological sensing layer comprises a biochemical sensing unit based on a flexible fiber substrate, the hormone level change is detected by using superparamagnetic nanoparticles and a specific aptamer compound, and non-invasive continuous monitoring is realized; the multi-element intervention execution layer comprises a temperature control unit, a bionic massage unit, a multi-dimensional acoustic unit and an aromatherapy release module, and supports multi-mode sleep intervention; the central processing system receives the physiological signals, dynamically generates an intervention strategy according to the sleep stage and the physiological state and drives the execution layer to operate, and closed-loop regulation and control are formed. Through multi-mode perception and cooperative intervention, accurate monitoring and self-adaptive adjustment of the sleep process are realized, and the sleep quality is improved.

Description

Adaptive intelligent sleep aid sleeping bag and sleep aid control method Technical Field

[0001] This application relates to intelligent sleep health devices, specifically to adaptive intelligent sleep aid sleeping bags and sleep aid control methods. Background Technology

[0002] Middle-aged women are a high-risk group for insomnia, with complex causes, primarily due to hormonal fluctuations around menopause that easily lead to hot flashes and night sweats, severely disrupting sleep structure. Furthermore, women in this age group often face multiple pressures from work and family (such as children's education and elderly parents' care), easily triggering anxiety and depression, leading to difficulty falling asleep and maintaining sleep. Fluctuations in body temperature at night, especially hot flashes, are a major factor causing nighttime awakenings. Joint and muscle aches and pains also increase.

[0003] In response to this, existing technologies include personal devices that combine massage functions with sleep aids. These devices integrate automatic massage elements into bedding such as pillows to improve the user's sleep quality. Such devices are typically equipped with a local central control processor capable of executing preset massage modes and time programs based on user-inputted commands.

[0004] In practical applications, physiological state also has a significant impact on sleep. To monitor physiological state, current control devices are connected to the internet to support remote monitoring. However, due to the lack of real-time detection capabilities for key physiological information of users, intervention strategies cannot be dynamically adjusted based on actual physical conditions. At the physiological level, fluctuations in estrogen and progesterone levels before and after menopause are one of the key factors affecting sleep. Currently, inferences about hormone changes mainly rely on predictive models or indirect indicators, which suffer from poor accuracy and lag in response, making it difficult to truly reflect individual hormonal dynamics. Although existing devices contain basic control logic, their operating modes depend on pre-set parameters and cannot adaptively adjust based on physiological changes during sleep. Furthermore, the integrated massage elements mostly use rigid rotating structures, which are insufficient to meet the comfort and safety requirements of the head and other sensitive areas of the body during sleep. Summary of the Invention

[0005] This application provides an adaptive sleep aid sleeping bag and control method based on flexible sensing and multimodal intervention, aiming to improve the problem that existing smart sleep devices cannot directly and continuously monitor the levels of key steroid hormones and lack the ability to personalize closed-loop regulation based on real-time physiological state, thus effectively improving the accuracy of regulation.

[0006] To achieve the above-mentioned objectives, the following technical solution is provided: An adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention is provided, comprising: the sleeping bag includes a sleeping bag body, the sleeping bag body integrating a physiological sensing layer, a multimodal intervention execution layer, and a central processing system; the physiological sensing layer is used to collect physiological signals, including a biochemical sensing unit; the biochemical sensing unit includes a flexible fiber substrate and a sensing functional layer immobilized on its surface, the sensing functional layer being composed of a complex of superparamagnetic nanoparticles and aptamers that specifically bind to target steroid hormones, used to detect the binding of hormones and aptamers... The changes in the Brownian relaxation time of nanoparticles caused by the combination enable non-invasive and continuous monitoring of the user's hormone levels; the multi-modal intervention execution layer includes a temperature control unit, a bionic massage unit, a multi-dimensional acoustic unit, and an aromatherapy release module, used to implement multimodal sleep intervention; the central processing system is electrically connected to the physiological sensing layer and the multi-modal intervention execution layer respectively, and is configured to receive the physiological signals, determine the user's current sleep stage and physiological state based on the physiological signals, determine the corresponding intervention strategy based on the current sleep stage and physiological state, and drive the multi-modal intervention execution layer to execute the corresponding intervention measures based on the intervention strategy.

[0007] In some embodiments, the biochemical sensing unit further includes a temperature reference subunit, which is an unmodified bare magnetic nanoparticle fiber that forms a differential pair with the sensing fiber for detection, and is used to monitor the temperature change of the sensing interface in real time and output a reference signal.

[0008] In some embodiments, the central processing system is configured to run a dynamic compensation and signal processing algorithm that, based on a reference signal output by a temperature reference subunit and a pre-calibrated temperature interference model, separates the temperature-affected component from the original mixed signal and performs differential compensation to obtain an interference-resistant biosignal signal.

[0009] In some embodiments, the temperature control unit includes an active temperature regulating material and a phase change material. The active temperature regulating material is provided with a micro-semiconductor cooler and a heating material. The active temperature regulating material is doped into the phase change material. The temperature control unit is distributed in the back, chest, or neck area. The bionic massage unit includes a flexible airbag array and a vibration motor. The bionic massage unit is positioned at at least one of the following acupoints: shoulder and neck, lumbosacral region, Shenshu acupoint, or Mingmen acupoint, to provide a gentle pressing or deep kneading mode. The multidimensional acoustic unit includes a directional speaker and a bone conduction unit. The bone conduction unit is used to play guided meditation audio while collecting skull vibration signals to assist in respiratory monitoring.

[0010] In some embodiments, the sleeping bag body is a multi-layer composite structure, comprising, from the inside out, a physiological sensing layer, a temperature control layer, a thermal insulation and buffer layer, and an outer fabric layer.

[0011] In some implementations, the central control module is also configured to execute a synchronous time-division multiplexing control strategy, enabling magnetic relaxation measurement and disabling the temperature control unit during odd-numbered time periods, and enabling the temperature control unit and disabling magnetic measurement during even-numbered time periods.

[0012] In some implementations, the intervention strategy includes the following triggering conditions and corresponding interventions: when an increase in pre-sleep skin conductance and a decrease in heart rate variability are detected, meditation audio, shoulder and neck massage, and aromatherapy release are activated; when a delayed transition from light sleep to deep sleep is identified, lumbosacral soothing massage is activated; when estradiol concentration decreases significantly and is accompanied by an increase in body surface temperature, chest cooling and alpha wave audio output are activated; when the body movement index exceeds a threshold, circulatory compression of the limb airbags is activated.

[0013] The above-mentioned adaptive sleep-aiding sleeping bag based on flexible sensing and multimodal intervention provides a sleep-aiding control method, including: collecting the user's physiological signals, including hormone levels, heart rate variability, respiratory rate, skin conductance, body surface temperature, and body movement data; inputting the physiological signals into a multimodal data fusion sleep staging algorithm, analyzing them based on a sliding time window, and outputting the current sleep stage determination result; matching corresponding intervention measures according to the sleep stage determination result and the presence of physiological abnormalities; and driving at least one of the temperature control unit, bionic massage unit, multidimensional acoustic unit, and aromatherapy release module to execute the matched intervention measures to achieve dynamic closed-loop control.

[0014] In some implementations, while performing temperature control intervention, the temperature change information of the sensing interface is acquired in real time through the temperature reference subunit, and the biochemical sensing signal is dynamically compensated in combination with the pre-calibrated temperature interference model to eliminate the interference caused by temperature control.

[0015] In some implementations, magnetic relaxation measurement and temperature control are alternately operated using a synchronous time-division multiplexing method: magnetic relaxation measurement is performed and temperature control is paused during the first time period of a time cycle, and temperature control is performed and magnetic measurement is paused during the second time period, utilizing the thermal inertia of the phase change material to maintain temperature stability. Beneficial effects

[0016] This invention relates to an adaptive smart sleep aid sleeping bag that achieves non-invasive, continuous monitoring of steroid hormones such as estrogen and progesterone through a smart fabric combining a specific fit with superparamagnetic nanoparticles. Compared to existing technologies that rely on predictive models or indirect physiological indicators to infer hormone levels, this approach directly responds to target hormone molecules diffused from the skin surface, significantly improving the accuracy and real-time nature of monitoring. This mechanism avoids misjudgments and delays caused by individual differences in traditional methods, providing a reliable basis for subsequent personalized interventions. Attached Figure Description

[0017] Figure 1 is a flowchart of the adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention in Embodiment 2 of the present invention, which also includes a temperature reference subunit. Detailed Implementation

[0018] To make the technical problems, technical solutions, and beneficial effects solved by this application clearer, the following detailed description is provided in conjunction with embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. Embodiment 1

[0019] This application provides an adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention, including a sleeping bag body. The sleeping bag body integrates a physiological sensing layer, a multimodal intervention execution layer, and a central processing system. The physiological sensing layer includes a biochemical sensing unit, which comprises a flexible fiber substrate and a sensing functional layer fixed on its surface. The sensing functional layer is composed of a complex of superparamagnetic nanoparticles and aptamers that specifically bind to target steroid hormones. It is used to achieve non-invasive continuous monitoring of the user's hormone levels by detecting changes in the Brownian relaxation time of nanoparticles caused by the binding of hormones and aptamers. The multimodal intervention execution layer includes a temperature control unit, a biomimetic massage unit, a multidimensional acoustic unit, and an aromatherapy release module, used to implement multimodal sleep intervention. The central processing system is electrically connected to the physiological sensing layer and the multimodal intervention execution layer and is configured to receive physiological signals collected by the physiological sensing layer, determine the user's current sleep stage and physiological state based on the physiological signals, determine the corresponding intervention strategy based on the current sleep stage and physiological state, and drive the multimodal intervention execution layer to execute the corresponding intervention measures based on the intervention strategy.

[0020] Compared to existing smart bedding that relies solely on heating functions or indirect physiological indicators to infer sleep status, this application integrates a biochemical sensing unit that can directly detect the concentration of steroid hormones such as estrogen and progesterone diffused on the skin surface. This enables non-invasive, continuous, and high-precision monitoring of key biomarkers, avoiding misjudgments and response delays caused by individual differences. It significantly improves the accuracy of sleep status recognition and the basic reliability of personalized intervention.

[0021] At the same time, it achieves the coordinated operation of multiple intervention methods such as temperature control, massage, acoustics and fragrance on the same device, breaking through the technical limitations of traditional products with single function and lack of closed-loop control capability.

[0022] The sleeping bag body has a five-layer composite structure, consisting of a physiological sensing layer, a temperature control layer, a thermal insulation buffer layer, and an outer fabric layer from the inside out. For example, the physiological sensing layer is 0.8 mm thick and uses silver-plated nylon 6 conductive fibers as the signal transmission bus, interwoven with sensing fibers to form a detection network with an array density of 20 points / cm². The sensing fiber substrate is made of polyethylene terephthalate (PET) microfiber felt with a fiber diameter of 2 μm, a porosity of 85%, and a thickness of 0.5 mm. Its surface is immobilized with an aptamer-magnetic nanoparticle composite. A bare magnetic nanoparticle reference fiber without aptamer modification is embedded at regular intervals to form a differential measurement pair, ensuring temperature field consistency ΔT < 0.1℃.

[0023] The temperature control layer is 2.5mm thick. Active temperature-regulating materials are distributed in the chest, back, and neck areas. Built-in micro Peltier chips (a type of micro-semiconductor cooler), measuring 4×4×2mm, with a maximum cooling power of 0.5W, are distributed in a 15mm grid. A 0.07mm thick graphite heat dissipation film with a thermal conductivity of 1500W / mK is adhered to the chip surface for uniform heat conduction. The phase change material is n-octadecane microcapsules with a phase change point of 28℃, a latent heat of 240J / g, and is mixed with addition-cured liquid silica gel. It has a Shore hardness of 30A, and the phase change material (PCM) mass fraction is 35%.

[0024] The limb areas are equipped with a passive temperature-regulating layer containing only phase change material. The thermal insulation buffer layer is a 3mm thick aerogel felt with a thermal conductivity of 0.018W / mK. The outer fabric is waterproof and breathable nylon 66, with a hydrostatic pressure of 10kPa and a moisture permeability of 5000g / m².24h. The central processing system is based on a flexible PCB with a polyimide substrate, 0.15mm thick, and integrates an STM32L4 series low-power MCU, a magnetic relaxation measurement front-end, a 24-bit ADC (analog-to-digital converter), and a Bluetooth 5.2 module.

[0025] For example, the preparation process of biochemical sensing fiber is as follows: First, Fe3O4 nanocrystals are synthesized by high-temperature thermal decomposition. Using acetylacetone iron(III) as a precursor, the reaction is carried out at 260°C for 2 hours in a phenyl ether / oleic acid system to obtain Fe3O4 cores with a particle size of 15±2nm. Then, a 20nm thick SiO2 shell is coated by reverse microemulsion method to form Fe3O4@SiO2 core-shell structure. APTES (3-aminopropyltriethoxysilane) is added and refluxed in ethanol at 70°C for 4 hours to achieve amino functionalization. The final product has a Zeta potential of +28mV. The aminated nanoparticles were dispersed in MES (2-(N-morpholino)ethanesulfonic acid) buffer (pH 5.5), and EDC / NHS activator (where EDC is 1-ethyl-(3-dimethylaminopropyl)carbodiimide hydrochloride and NHS is N-hydroxysuccinimide) was added and shaken for 30 minutes. Then, it was coupled overnight at 4°C with a 43nt DNA aptamer modified with -NH2 at the 5′ end (sequence: 5′-NH2-AAAAAAAAAACGTCGATCGTGTGTGTGTGTGTGTGTGTGTGTGTGT-3′, dissociation constant Kd=12nM). After magnetic separation, the aptamer-magnetic nanoparticle complex was washed three times with phosphate-buffered saline (PBS) and blocked with 1% bovine serum albumin (BSA) at 37°C for 1 hour to obtain a final concentration of 5 mg / mL. The PET fiber felt was immersed in the above composite solution and ultrasonically assisted (40kHz) for 10 minutes to allow the nanoparticles to penetrate into the fiber to a depth of 50–100μm. Then, it was cross-linked with 0.5% glutaraldehyde vapor for 30 minutes to complete the covalent anchoring. The loading was controlled at 8–10mg nanoparticles / g fiber, corresponding to a aptamer density of about 2×10¹³ particles / cm².

[0026] The principle of magnetic relaxation detection is as follows: when 17β-estradiol binds to the aptamer, the rotational correlation time τ_R of the nanoparticles increases from 0.5 ns to 3.2 ns, causing the imaginary part of the AC magnetic susceptibility Δχ″ to decrease at the characteristic frequency f0 = 25 kHz. Calibration experiments were conducted at a constant temperature of 30℃. The sensing fiber was immersed in a 0–200 pg / mL 17β-estradiol standard solution, and the changes in Δχ″ were recorded and fitted using the Hill equation: Δχ″(C) = (0.12·C¹). 8 ) / (1.8+C¹). 8 This study established a quantitative relationship between hormone concentration and signal. The limit of detection (LOD) was 3.2 pg / mL, the linear range was 5–150 pg / mL, the response time was less than 5 minutes, and the specificity was greater than 98% (no cross-reactivity with progesterone and testosterone).

[0027] The above scheme enables non-invasive continuous monitoring of steroid hormones such as estrogen and progesterone. This is because the binding of the target hormone to the aptamer causes changes in the viscosity or steric hindrance of the microenvironment surrounding the superparamagnetic nanoparticles, leading to an increase in the rotational correlation time τ_R from 0.5 ns to 3.2 ns. This, in turn, causes a decrease in the imaginary part of the AC magnetic susceptibility Δχ″ at the characteristic frequency f0 = 25 kHz. This signal change is quantitatively related to the hormone concentration. This detection method is based on the principle of magnetic relaxation. A 17β-estradiol standard solution is calibrated at a constant temperature of 30℃, and the Hill equation Δχ″(C) = (0.12·C¹) is fitted. 8 ) / (1.8+C¹). 8 The detection limit is 3.2 pg / mL, the linear range is 5–150 pg / mL, the response time is less than 5 minutes, and the specificity is greater than 98% (no cross-reactivity with progesterone and testosterone). This solves the problems of poor accuracy and response lag caused by traditional techniques that rely on indirect indicators to infer hormone fluctuations. Example 2

[0028] Referring to Figure 1, the adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention provided in this embodiment of the application also includes a temperature reference subunit. The temperature reference subunit is a bare magnetic nanoparticle fiber without modification, which forms a differential pair with the sensing fiber for detection, used to monitor the temperature change of the sensing interface in real time and output a reference signal. By introducing a differential temperature reference structure, the sensing signal interference caused by temperature fluctuations can be accurately identified and separated when the temperature control function is running synchronously, thereby ensuring the stability and reliability of the main sensing channel in a dynamic thermal environment, and solving the problem of biochemical sensing drift or even failure caused by temperature control operation in the prior art.

[0029] The above scheme enables high-fidelity extraction of biochemical sensor signals under temperature control. The system first queries a pre-calibrated mapping table based on the reference channel signal and the measured value of the Pt1000 micro-thermal resistor to calculate the accurate temperature estimate T_est(t). Then, T_est(t) is input into the temperature interference model S_interf(t) = 0.8ΔT + 0.02ΔT² + 0.1(dT / dt) (where ΔT = T_est(t) - T_calib, T_calib = 30℃) to calculate the pure temperature interference component. Subsequently, S_interf(t) is subtracted from the original signal S_raw(t) to obtain the interference-resistant biosignal S_pure(t). Finally, a stable concentration estimate is output by combining the temperature-corrected Hill inversion model Kd_T = 15.0 × (1 + 0.04 × ΔT) with Kalman filtering (process noise Q = 0.01, measurement noise R = 0.25). Experimental data show that within a temperature variation range of ±10℃, temperature interference decreased from ±18 pg / mL to ±0.5 pg / mL, with an elimination rate of over 97%, enabling temperature control and sensing to operate collaboratively in the same time and space.

[0030] For example, the dynamic compensation and signal processing algorithm runs in the MCU (microcontroller) of the central processing system, executing once every 30 seconds. First, the signal output from the temperature reference subunit is read. This reference fiber contains only bare magnetic nanoparticles, without any aptamer modification, and is used to monitor the temperature change at the sensing interface in real time. Combined with the Pt1000 micro-thermal resistance measurement value, a pre-calibrated mapping table is consulted to calculate the accurate temperature estimate T_est(t). T_est(t) is input into a pre-established temperature disturbance model, expressed as S_interf(t) = 0.8ΔT + 0.02ΔT² + 0.1(dT / dt), where ΔT = T_est(t) - T_calib (T_calib = 30℃), used to calculate the pure temperature disturbance component S_interf(t). S_interf(t) is subtracted from the raw mixed signal S_raw(t) collected by the biochemical sensing unit to obtain the interference-resistant biosignal signal S_pure(t). S_pure(t) and T_est(t) are input into a concentration inversion model, which is built based on multidimensional calibration data. The model uses the Hill equation but introduces a temperature correction term: Kd_T = 15.0 × (1 + 0.04 × ΔT), to calculate the estimated target hormone concentration C_est(t). The final output is smoothed using a Kalman filter, resulting in process noise Q = 0.01 and measurement noise R = 0.25, effectively suppressing random fluctuations. Experiments show that within the range of ΔT = 10℃, temperature interference is reduced from ±18 pg / mL to ±0.5 pg / mL, with an elimination rate greater than 97%. Example 3

[0031] The adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention provided in this application embodiment also includes dynamic compensation and signal processing algorithms. Utilizing the reference signal output by the temperature reference subunit, combined with a pre-calibrated temperature interference model, the component affected by temperature interference is separated from the original mixed signal and differential compensation is performed to obtain anti-interference biosignal signals. This algorithm effectively eliminates thermal interference generated during temperature control through a differential compensation mechanism, enabling continuous active temperature control intervention without interrupting biochemical monitoring, thus resolving the contradiction that traditional devices must pause sensing to ensure temperature control accuracy. For example, to avoid interference from Peltier chip electromagnetic noise on magnetic measurements, the system adopts a synchronous time-division multiplexing control strategy: during odd seconds (0–29s), the magnetic relaxation measuring instrument operates, applying a 25kHz, 1mT excitation magnetic field, while the Peltier chip is in a dormant state; during even seconds (30–59s), the Peltier chip initiates PID (proportional-integral-derivative) temperature control, and the magnetic measuring instrument stops operating. By utilizing the thermal inertia of the phase change material layer (time constant τ = 180s), the temperature fluctuation is maintained at less than ±0.3℃ during the dormancy period, achieving seamless integration of temperature control and monitoring functions. Example 4

[0032] The temperature control unit of the adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention provided in this application embodiment includes an active temperature-regulating material and a passive temperature-regulating zone. The active temperature-regulating material is equipped with a miniature semiconductor cooler and a phase change material, and is distributed in the back, chest, or neck areas. The passive temperature-regulating zone is filled only with phase change material and is located in the limb areas. This zoned temperature control design takes into account both the rapid response capability of key areas and the thermal buffering needs of the limb areas. It not only meets the requirements for precise intervention in sudden situations such as hot flashes, but also reduces energy consumption and sudden local temperature changes through passive temperature regulation, thereby improving overall comfort and energy efficiency.

[0033] In this embodiment, the bionic massage unit includes a flexible airbag array and a vibration motor, positioned at at least one of the following acupoints: shoulder and neck, lumbosacral region, Shenshu acupoint, or Mingmen acupoint. Driven by a micro-pump and an array of solenoid valves, it provides gentle pressing or deep kneading modes. Compared to traditional rigid massage devices, this flexible massage structure offers superior fit and pressure distribution uniformity, enabling targeted muscle relaxation under low-noise conditions. It is particularly suitable for soothing interventions in easily fatigued areas such as the shoulder, neck, and lower back during sleep, enhancing safety and user experience. For example, the bionic massage unit is arranged in the shoulder and neck (C5-T1 segment), lumbosacral region (L4-S2 segment), Shenshu acupoint (BL23), and Mingmen acupoint (DU4). Each position is equipped with a flexible airbag array made of nylon 66 and TPU composite film with a thickness of 0.2mm. The size of a single airbag is 3×3cm, and the maximum pressure it can withstand is 50kPa. The drive system consists of a micro air pump with a size of 2×2×1.5cm and a noise level of less than 25dB, and an array of solenoid valves, which supports two working modes: 0.3Hz gentle pressing (for hot flash intervention) and 1Hz deep kneading (for muscle relaxation).

[0034] In this embodiment, the multi-dimensional acoustic unit includes a directional speaker and a bone conduction unit. The bone conduction unit is used to play guided meditation audio while simultaneously acquiring skull vibration signals to assist in respiratory monitoring. This integrated acoustic design achieves the dual functions of private acoustic intervention and non-contact respiratory monitoring. The directional speaker avoids disturbing the bed partner, while the bone conduction unit can both output sleep-inducing audio and conversely acquire skull micro-vibrations caused by breathing. Respiratory frequencies are extracted through adaptive filtering with an accuracy of ±1 bpm, enriching multimodal data input without the need for additional sensors. For example, the multi-dimensional acoustic unit includes a directional speaker and a bone conduction unit built into the headrest; the directional speaker has a frequency response range of 100Hz–8kHz, a maximum sound pressure level of 75dB (@30cm), and a beamwidth of ±30° to avoid disturbing the bed partner. The directional speaker plays white noise such as natural wind sounds or stream sounds. The bone conduction unit uses a piezoelectric ceramic oscillator with a resonant frequency of 500Hz, which is attached to the temporal region of the skull to play guided meditation audio of theta waves (4~7Hz) and alpha waves (8~12Hz). At the same time, it collects the micro-vibrations of the skull (0.1~0.5Hz) caused by breathing in the reverse direction, and extracts the breathing frequency through adaptive filtering with an accuracy of ±1bpm.

[0035] In this embodiment, the sleeping bag body has a multi-layer composite structure, comprising, from the inside out, a skin contact layer, a physiological sensing layer, a temperature control execution layer, a thermal insulation buffer layer, and an outer fabric layer, with a total thickness of 8-10 mm. This composite structure integrates five functional layers within a limited thickness, balancing wearing comfort, functional integrity, and environmental adaptability. The overall bending stiffness of the sleeping bag is less than 0.1 Nm, its weight is less than 2.5 kg, and it can withstand 50 machine washes without a decrease in sensitivity of more than 15%, solving the problems of traditional smart bedding being stiff, difficult to clean, and inconvenient to store. Example 5

[0036] This application also provides an adaptive intelligent sleep aid control system, including a physiological sensing module, an intervention execution module, and a central control module. The physiological sensing module is used to collect the user's hormone levels, heart rate variability, respiratory rate, skin conductance, body surface temperature, and body movement data. The intervention execution module includes a temperature control unit, a bionic massage unit, a multidimensional acoustic unit, and an aromatherapy release module, used to implement multimodal sleep intervention. The central control module is configured to run a multimodal data fusion sleep staging algorithm and a female-specific AI intervention engine. The multimodal data fusion sleep staging algorithm outputs the current sleep stage determination result based on multiple physiological signals within a sliding time window. The female-specific AI intervention engine matches and executes corresponding intervention measures from a strategy library according to the sleep stage and abnormal physiological state, forming a closed-loop regulation. This system integrates hormones as a key biomarker with other multimodal physiological parameters, uses a lightweight CNN model to achieve high-precision sleep staging recognition (Kappa coefficient reaches 0.82), and constructs an AI intervention strategy library based on female-specific physiological rhythms, realizing precise and dynamic closed-loop regulation of typical sleep disorders in perimenopausal women. The central processing system runs a multimodal data fusion sleep staging algorithm. Input signals include heart rate variability (HRV) within a 30-second sliding window (calculated by measuring the RMSSD and LF / HF ratio from pulse waves acquired via piezoelectric fiber sensors), respiratory rate (extracted from bone conduction units, range 8–25 bpm), triaxial accelerometer body motion signals (sampling rate 50 Hz, body motion index <5 indicates a stable phase), skin conductance (measured via conductive fiber mesh, rising >2 μS during hot flashes), and hormone levels (output from magnetic relaxation sensors). The classification model is a lightweight CNN structure, containing three convolutional layers and one fully connected layer, with fewer than 50k parameters. It was trained on a 2000-hour labeled sleep dataset and outputs five labels: wakefulness (W), N1, N2, N3, and REM sleep stages. Compared with the gold standard polysomnography (PSG), the Kappa coefficient reached 0.82. The female-specific AI intervention engine invokes intervention measures from the strategy library based on the current sleep stage and physiological state: When a significant increase in pre-sleep skin conductance and a decrease in HRV are detected, meditation audio playback, gentle massage of the neck and shoulders, and aromatherapy release are initiated for 15 minutes, with the priority being acoustics > massage > aromatherapy; When a delayed transition from light sleep (N2) to deep sleep (N3) is identified, a lumbar soothing massage program (0.3Hz, 20kPa) is triggered for 10 minutes, executed independently; When a decrease in estradiol concentration of more than 50% and a rise in local body surface temperature of more than 1°C are detected, a hot flash event is identified, and chest area cooling and alpha wave audio output are initiated, continuing until hormone levels recover, with the priority being temperature control > acoustics; When the body movement index is greater than 10, circulatory massage of the limb airbags is initiated for 5 minutes, with the priority being massage > temperature control.If the target physiological indicators do not improve within 5 minutes (e.g., skin temperature does not decrease), the system automatically upgrades the intervention intensity (e.g., the Peltier power is increased from 50% to 100%), forming a closed-loop feedback regulation mechanism. Example 6.

[0037] The central control module of the adaptive intelligent sleep aid control system provided in this application embodiment is further configured to execute a synchronous time-division multiplexing control strategy. Magnetic relaxation measurement is enabled and the temperature control unit is disabled during odd-numbered time periods, while the temperature control unit is enabled and magnetic measurement is disabled during even-numbered time periods. This utilizes the thermal inertia of the phase change material to maintain temperature stability and avoid electromagnetic interference. This time-division multiplexing mechanism effectively avoids interference from Peltier chip electromagnetic noise on high-sensitivity magnetic relaxation measurement, achieving uninterrupted collaborative operation of the two core functions without increasing electromagnetic shielding costs, thus improving system integration and reliability.

[0038] In this embodiment, the central control system is further configured to execute a strategy library for a female-specific AI intervention engine, including the following triggering conditions and corresponding intervention measures: when an increase in pre-sleep skin conductance and a decrease in heart rate variability are detected, meditation audio, shoulder and neck massage, and aromatherapy release are activated; when a delayed transition from light sleep to deep sleep is identified, lumbosacral soothing massage is activated; when estradiol concentration decreases significantly and is accompanied by an increase in body surface temperature, chest cooling and alpha wave audio output are activated; when the body movement index exceeds a threshold, circulatory airbag massage of the limbs is activated. This strategy library stores refined intervention logic for typical female sleep disorder scenarios, and can automatically match the optimal intervention combination according to different physiological states, which is significantly better than fixed programs or manual control modes, improving the scientific nature and effectiveness of the intervention.

[0039] In this embodiment, the female-specific AI intervention engine is also configured to automatically upgrade the intervention intensity if the target physiological indicators do not improve within a set time. This mechanism achieves true dynamic feedback adjustment, breaking through the limitations of traditional open-loop control. It can dynamically adjust the output intensity according to the actual intervention effect, ensuring the rapid recovery of physiological homeostasis. Clinical data shows that after 30 days of use, the proportion of deep sleep increased to 26.8%, and the overall sleep efficiency increased to 83%.

[0040] In addition, a mobile application is provided that generates visual reports, highlighting the frequency, duration, sleep structure ratio, and stress index of nighttime hot flashes. It offers one-click selection of modes such as "Menstrual / Menopausal Mode," "Stress Reduction Mode," and "Deep Sleep Mode." It also provides health knowledge education and community support functions specifically for middle-aged women. Example 7

[0041] This application also provides an adaptive intelligent sleep aid control method, including collecting the user's physiological signals, including hormone levels, heart rate variability, respiratory rate, skin conductance, body surface temperature, and body movement data; inputting the physiological signals into a multimodal data fusion sleep staging algorithm, analyzing them based on a sliding time window, and outputting the current sleep stage determination result; based on the sleep stage determination result and the presence of any abnormal physiological states, a female-specific AI intervention engine matches corresponding intervention measures from a preset strategy library; driving at least one of the temperature control unit, bionic massage unit, multidimensional acoustic unit, and aromatherapy release module to execute the matched intervention measures, achieving dynamic closed-loop regulation. This method ensures that the intervention strategy is always synchronized with the current sleep process by sliding and updating the analysis window every 30 seconds, integrating the key variable of hormones, and making up for the shortcomings of relying solely on traditional parameters to determine the sleep stage, especially suitable for people whose sleep quality is significantly affected by hormone fluctuations. The method collects the user's physiological signals, including heart rate variability, respiratory rate, skin conductance, body surface temperature, body movement data, and hormone levels. These signals are input into a multimodal data fusion sleep staging algorithm, analyzed once every set time window, and the current sleep stage determination result is output. Based on the sleep stage assessment results and abnormal physiological states, a female-specific AI intervention engine matches and executes corresponding intervention measures from the strategy library to achieve dynamic closed-loop regulation.

[0042] In this embodiment, as shown in Figure 1, the adaptive intelligent sleep aid control method, while performing temperature control intervention, acquires real-time temperature change information of the sensing interface through a temperature reference subunit. It then combines this information with a pre-calibrated temperature interference model to dynamically compensate the biochemical sensing signal, thereby eliminating interference caused by temperature control. This compensation process is executed every 30 seconds, reading the reference channel signal and Pt1000 temperature data, calculating T_est(t); substituting this into the temperature interference model to obtain S_interf(t); subtracting this component from S_raw(t) to obtain S_pure(t); and finally combining this with a temperature-corrected concentration inversion model to output C_est(t). This ensures that reliable hormone concentration estimates can be obtained even under drastic temperature changes, supporting long-term stable closed-loop regulation.

[0043] In this embodiment, a synchronous time-division multiplexing method is used to alternately run magnetic relaxation measurement and temperature control operation: magnetic relaxation measurement is performed and temperature control is paused during the first time period of a time cycle, and temperature control is performed and magnetic measurement is paused during the second time period, utilizing the thermal inertia of the phase change material to maintain temperature stability. Specifically, the time cycle is set to 60 seconds, with the first 30 seconds used for magnetic relaxation measurement (excitation parameters: 25kHz, 1mT), and the last 30 seconds used for temperature control operation (PID control). Since the phase change material has a thermal time constant of 180 seconds, the temperature fluctuation within the 30-second measurement interval is less than ±0.3℃, meeting the requirements for sleep comfort. High-precision magnetic detection and efficient temperature control can coexist without additional electromagnetic shielding.

[0044] In this application, unless otherwise expressly defined, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0045] The terms “first,” “second,” “third,” “fourth,” etc., in this application (if present) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0046] Preferred embodiments of the invention will now be described in more detail. While preferred embodiments of the invention have been shown, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0047] The terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a” and “the” as used in this invention and the appended claims are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0048] It should be understood that although the terms "first," "second," "third," etc., may be used in this invention to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this invention, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, features defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0049] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention, characterized in that, include: The sleeping bag includes a sleeping bag body, which integrates a physiological sensing layer, a multi-intervention execution layer, and a central processing system. The physiological sensing layer is used to collect physiological signals, including a biochemical sensing unit. The biochemical sensing unit comprises a flexible fiber substrate and a sensing functional layer immobilized on its surface. The sensing functional layer is composed of a complex of superparamagnetic nanoparticles and aptamers that specifically bind to target steroid hormones. This complex is used to achieve non-invasive continuous monitoring of the user's hormone levels by detecting changes in the Brownian relaxation time of the nanoparticles caused by hormone binding to the aptamers. The multi-modal intervention execution layer includes a temperature control unit, a biomimetic massage unit, a multi-dimensional acoustic unit, and an aromatherapy release module, used to implement multimodal sleep intervention. The central processing system is electrically connected to both the physiological sensing layer and the multi-modal intervention execution layer, and is configured to receive the physiological signals, determine the user's current sleep stage and physiological state based on the physiological signals, determine corresponding intervention strategies based on the current sleep stage and physiological state, and drive the multi-modal intervention execution layer to execute corresponding intervention measures based on the intervention strategies.

2. The adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention according to claim 1, characterized in that, The biochemical sensing unit also includes a temperature reference subunit, which is an unmodified bare magnetic nanoparticle fiber that forms a differential pair with the sensing fiber for detection, and is used to monitor the temperature change of the sensing interface in real time and output a reference signal.

3. The adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention according to claim 2, characterized in that, The central processing system is configured to run a dynamic compensation and signal processing algorithm. The algorithm separates the temperature-affected components from the original mixed signal based on the reference signal output by the temperature reference subunit and a pre-calibrated temperature interference model, and performs differential compensation to obtain interference-resistant biosignal signals.

4. The adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention according to claim 1, characterized in that, The temperature control unit includes an active temperature regulating material and a phase change material. The active temperature regulating material is equipped with a micro-semiconductor cooler and a heating material. The active temperature regulating material is doped into the phase change material. The temperature control unit is distributed in the back, chest, or neck area. The bionic massage unit includes a flexible airbag array and a vibration motor. The bionic massage unit is positioned at at least one of the following acupoints: shoulder and neck, lumbosacral region, Shenshu acupoint, or Mingmen acupoint, to provide gentle pressing or deep kneading modes. The multidimensional acoustic unit includes a directional speaker and a bone conduction unit. The bone conduction unit is used to play guided meditation audio while collecting skull vibration signals to assist in respiratory monitoring.

5. The adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention according to claim 1, characterized in that, The sleeping bag body has a multi-layer composite structure, which includes, from the inside out, a physiological sensing layer, a temperature control layer, a heat insulation and buffer layer, and an outer fabric layer.

6. The adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention according to claim 1, characterized in that, The central control module is also configured to execute a synchronous time-division multiplexing control strategy, enabling magnetic relaxation measurement and disabling the temperature control unit during odd-numbered time periods, and enabling the temperature control unit and disabling magnetic measurement during even-numbered time periods.

7. The adaptive sleep aid sleeping bag based on flexible sensing and multimodal intervention according to claim 1, characterized in that, The intervention strategy includes the following triggering conditions and corresponding intervention measures: when an increase in pre-sleep skin conductance and a decrease in heart rate variability are detected, meditation audio, shoulder and neck massage, and aromatherapy release are activated; when a delay in the transition from light sleep to deep sleep is identified, soothing pressure on the lumbosacral region is activated; when estradiol concentration decreases significantly and is accompanied by an increase in body surface temperature, chest cooling and alpha wave audio output are activated; when the body movement index exceeds a threshold, circulatory pressure on the limb airbags is activated.

8. The sleep-aid control method for an adaptive sleep-aid sleeping bag based on flexible sensing and multimodal intervention as described in any one of claims 1 to 7, characterized in that, include: Collect users' physiological signals, including hormone levels, heart rate variability, respiratory rate, skin conductance, body surface temperature, and body movement data; The physiological signals are input into a multimodal data fusion sleep staging algorithm, analyzed based on a sliding time window, and the current sleep stage determination result is output. According to the sleep stage determination result and whether there is a physiological abnormality, corresponding intervention measures are matched. At least one of the temperature control unit, bionic massage unit, multidimensional acoustic unit and aromatherapy release module is driven to execute the matched intervention measures to achieve dynamic closed-loop regulation.

9. The sleep-aid control method for an adaptive sleep-aid sleeping bag based on flexible sensing and multimodal intervention according to claim 8, characterized in that, While performing temperature control intervention, the temperature change information of the sensing interface is acquired in real time through the temperature reference subunit. Combined with the pre-calibrated temperature interference model, the biochemical sensing signal is dynamically compensated to eliminate the interference caused by temperature control.

10. The sleep-aid control method for an adaptive sleep-aid sleeping bag based on flexible sensing and multimodal intervention according to claim 8, characterized in that, The magnetic relaxation measurement and temperature control operation are alternately operated using a synchronous time-division multiplexing method: magnetic relaxation measurement is performed and temperature control is paused during the first time period of a time cycle, and temperature control is performed and magnetic measurement is paused during the second time period, using the thermal inertia of the phase change material to maintain temperature stability.

Images