Sleep-aiding pillow

Abstract

The invention discloses a sleep-aiding pillow, and relates to the field of sleep science and technology, personalized production technology and bioengineering, the sleep-aiding pillow comprises a pillow body, a biosensor, a microcontroller and a vibration device; the biosensor and the vibration device are both connected with the microcontroller; the microcontroller and the vibration device are arranged in the pillow body; the biosensor is used for collecting the heart rate, the respiratory rate and the activity level of a user; the microcontroller is used for controlling the vibration frequency of the vibration device according to the vibration frequency of the vibration device and the heart rate, the respiratory rate and the activity level of the user so as to assist the user in sleeping and maintain the user in a deep sleep state; the pillow body is obtained by printing through the 3D printing technology according to the head and neck picture of the user, and the pillow helps the user to enter deep sleep more quickly and keep the deep sleep state.

Description

Technical Field

[0001] The present application relates to the fields of sleep technology, personalized production technology and bioengineering, and in particular to a sleep-aiding pillow. Background Art

[0002] Sleep is an important part of human health and overall quality of life, but many people face challenges in getting high-quality sleep. Common problems include insomnia, poor sleep quality, and frequent waking up at night. These problems may lead to fatigue, lack of energy, memory loss, and even more serious health problems. For people who have suffered a stroke, ruptured aneurysm, or cognitive impairment, adequate sleep and neural repair during sleep are extremely important.

[0003] Some companies and research institutions have launched devices that use noise or music to promote sleep. In addition, there are some products such as smart pillows, which use internal sensors to monitor the pressure points of the user's head and neck and automatically adjust the hardness and shape of the pillow to provide more comfortable support, but the final sleep-aiding effect of these devices and smart pillows is not satisfactory. Summary of the invention

[0004] The purpose of the present application is to provide a sleep-aiding pillow that can help a user enter a deep sleep faster and maintain a deep sleep state.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, a sleeping aid pillow comprises:

[0007] A pillow body, a biosensor, a microcontroller and a vibration device; the biosensor and the vibration device are both connected to the microcontroller; the microcontroller and the vibration device are arranged inside the pillow body; the biosensor is used to collect the user's heart rate, breathing frequency and activity level; the microcontroller is used to control the vibration frequency of the vibration device according to the vibration frequency of the vibration device, the user's heart rate, breathing frequency and activity level, so as to assist the user to fall asleep and maintain the user in a deep sleep state; the pillow body is printed using 3D printing technology based on a photo of the user's head and neck.

[0008] Optionally, the biosensor is an embedded biosensor.

[0009] Optionally, the microcontroller comprises:

[0010] The control module is used to input the vibration frequency of the vibration device at the current moment, the user's heart rate, breathing rate and activity level into the frequency control model to obtain the regulation strategy of the vibration device, and control the vibration frequency of the vibration device at the current moment according to the regulation strategy; the regulation strategy is to increase the vibration frequency, reduce the vibration frequency or keep it unchanged; the frequency control model is trained by using a deep reinforcement learning algorithm.

[0011] Optionally, the deep reinforcement learning algorithm is a DQN algorithm.

[0012] Optionally, the pillow body construction process specifically includes:

[0013] Obtaining a photograph of the user's head and neck;

[0014] Use a deep learning model to perform face recognition on the user's head and neck photos to obtain the position of the user's head;

[0015] The user's head and neck photo is processed using a contour model to obtain a 3D point cloud of the user's head and neck; the contour model is obtained by training the 3DAttriFlow model;

[0016] Obtaining the size and shape of the user's neck based on the user's head and neck photo;

[0017] Building a 3D model based on the position of the user's head, the 3D point cloud of the user's head and neck, and the size and shape of the user's neck;

[0018] The 3D model is printed using 3D printing technology to obtain the pillow body.

[0019] Optionally, the vibration device comprises:

[0020] A vibration motor, a shock absorbing layer and a control unit, wherein the shock absorbing layer wraps the vibration motor, the vibration motor is connected to the control unit, and the control unit is connected to the control module.

[0021] Optionally, the sleeping-aiding pillow further comprises: a battery and a charging module, the battery is connected to the charging module, and the battery is connected to the vibration device.

[0022] Optionally, the sleep-aiding pillow further includes: a pillowcase.

[0023] According to the specific embodiments provided in this application, this application has the following technical effects:

[0024] The present application provides a sleeping pillow. The devices in the related art that use noise or music to promote falling asleep often use preset music or noise patterns, but sound stimulation is not friendly to sleep. In addition, although the smart pillows in the related art can adjust their shape and hardness, their effects often vary from person to person, and not everyone can find the support method that best suits them. In addition, these devices often do not take into account the user's sleep cycle and cannot provide appropriate stimulation at the right time to promote deep sleep. The present application promotes sleep through vibration. The pillow body is printed using 3D printing technology based on a photo of the user's head and neck to make it more suitable for the user. The vibration frequency of the vibration device is controlled according to the user's heart rate, breathing rate and activity level, and appropriate stimulation is provided at the right time to promote deep sleep, assisting the user to fall asleep quickly and maintaining the user in a deep sleep state. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 A schematic diagram of the structure of a sleep-aiding pillow provided in one embodiment of the present application;

[0027] Figure 2 Schematic diagram of the structure of the 3D AttriFlow model. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0029] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0030] 40hz stimulation has been widely used in clinical research to alleviate cognitive impairment. The scientific principle of this potential therapeutic effect may be that frequency stimulation can affect neural plasticity, which is the ability of the brain to change its structure and function to adapt to environmental changes. When the brain is subjected to continuous 40Hz frequency stimulation, it may induce a plasticity phenomenon called "frequency following response." This is a way for the brain to change itself to better respond to this stimulation. In this process, the connection between neurons may be strengthened, and new neural networks may be formed, thereby improving the brain's cognitive function. However, this effect is currently often achieved through two stimulation methods, sound and light, but sound and light stimulation is not friendly to sleep, and compared with additional equipment, pillows are more convenient. Based on this, an embodiment of the present application provides a sleep-aid pillow, such as Figure 1 As shown, the sleep-aiding pillow includes: a pillow body, a biosensor, a microcontroller (a microcontroller is a small computer that integrates a processor, a memory and various interfaces, and can run pre-written programs to process data and control equipment) and a vibration device; the biosensor and the vibration device are both connected to the microcontroller; the microcontroller and the vibration device are arranged inside the pillow body; the biosensor is used to collect the user's heart rate, breathing frequency and activity level; the microcontroller is used to control the vibration frequency of the vibration device according to the vibration frequency of the vibration device, the user's heart rate, breathing frequency and activity level, so as to assist the user to fall asleep and maintain the user in a deep sleep state; the pillow body is printed using 3D printing technology based on a photo of the user's head and neck. This method ensures that the shape of the pillow completely fits the shape of the user's head and neck to provide the best support.

[0031] In another exemplary embodiment of the present application, the biosensor is an embedded sensor that monitors the user's heart rate, breathing rate and body activity, and collects physiological data in real time.

[0032] In another exemplary embodiment of the present application, the microcontroller includes:

[0033] The control module is used to input the vibration frequency of the vibration device at the current moment, the user's heart rate, breathing rate and activity level into the frequency control model to obtain the regulation strategy of the vibration device, and control the vibration frequency of the vibration device at the current moment according to the regulation strategy to obtain the vibration frequency of the vibration device at the next moment; the regulation strategy is to increase the vibration frequency, reduce the vibration frequency or keep it unchanged; the frequency control model is obtained by training with a deep reinforcement learning algorithm.

[0034] In another exemplary embodiment of the present application, a deep reinforcement learning algorithm is used for training, specifically a deep Q-network (DQN) is used as the core algorithm to achieve personalized and adaptive vibration frequency adjustment for the user. The user's physiological state (including heart rate, breathing rate, activity level, etc.) is monitored in real time through biosensors. The system dynamically optimizes the frequency of the vibration device based on these data to help the user quickly enter and maintain deep sleep. The specific process is a well-known process, and the general steps are:

[0035] State Space Definition (StateSpace): The user's current physiological state and environmental parameters are defined as the system's state space, including heart rate, breathing rate, activity level, and current vibration frequency. The heart rate is obtained through the bracelet, and the breathing and activity level are obtained by the vibration sensor built into the pillow, expressed as:

[0036] S t = {heart rate (HR t ), respiratory rate (BR t ), activity level t ), the current vibration frequency (Freq t )}.

[0037] Action Space Definition (ActionSpace): The action space of the system includes different options for adjusting the vibration frequency, such as increasing the vibration frequency, decreasing the vibration frequency, or keeping it unchanged: expressed as: A = {increase the vibration frequency, decrease the vibration frequency, keep it unchanged}.

[0038] Reward Function Design (RewardFunction): The reward function is used to evaluate the effect of each vibration adjustment, with the goal of improving sleep quality. The design is as follows:

[0039] R t+1 =α·Sleep quality score-β·|ΔHR t |-γ·|ΔBR t |

[0040] Among them, α, β, and γ are weight coefficients, and ΔH and ΔB represent the changes in heart rate and respiratory rate, respectively. t is the reward function of the previous stage.

[0041] Deep Q Network (DQN) model training: The DQN model approximates the Q value function through a neural network. The input layer of the network receives the current state S t , the output layer is the Q value Q(s,a;θ) corresponding to each possible action. Through experience replay (ExperienceReplay) and fixed target network (TargetNetwork), the system continuously optimizes the strategy to maximize the long-term cumulative reward.

[0042] Action selection strategy: The ε-greedy strategy is used to balance exploration and utilization during the adjustment process, ensuring that the system can make adaptive adjustments to state changes while learning the optimal strategy. That is, the vibration frequency adjustment direction is selected according to the model prediction with a probability of ε, and the adjustment is made randomly with a probability of 1-ε, and ε gradually increases during the adjustment process.

[0043] By using a deep Q network (DQN) to achieve intelligent control of the vibration device, the system can dynamically adjust the vibration frequency according to the user's specific physiological state, providing a more refined and personalized sleep assistance effect. This adaptive adjustment method based on reinforcement learning significantly improves the effect of the pillow in helping users fall asleep quickly and maintain a deep sleep state.

[0044] In another exemplary embodiment of the present application, the pillow body construction process specifically includes:

[0045] Obtain photos of the user's head and neck (pictures of the user's head and neck from the front, back, left, and right directions).

[0046] A deep learning model is used to perform face recognition on photos of the user's head and neck to obtain the position of the user's head.

[0047] The user's head and neck photo is processed using a contour model to obtain a 3D point cloud of the user's head and neck (a curved surface including the head and neck); the contour model is obtained by training the 3D AttriFlow model.

[0048] The size and shape of the user's neck are obtained based on the user's head and neck photos. The size and shape of the user's neck are obtained by comparing the photos uploaded by the user with the standard scale pictures to determine the relative size of the neck and head.

[0049] A 3D model is constructed based on the position of the user's head, the 3D point cloud of the user's head and neck, and the size and shape of the user's neck.

[0050] The 3D model is printed using 3D printing technology to obtain the pillow body.

[0051] The 3D AttriFlow model is used to convert 2D images into 3D point clouds. The input of this model is multi-angle photos and some user input (such as whether the person has long hair, facial expressions, etc.), and the output is a surface (such as Figure 2 The training process is as follows:

[0052] First, multi-angle photos and 3D reconstruction data (a curved surface including the head and neck) are obtained. This is the data acquisition step.

[0053] Then the data is preprocessed. The first step is to mark the key spatial points (landmark annotation); the second step is to cut out the head area separately; the third step is to align the spatial points to the middle of the area (3D Alignment); the fourth step is to flatten the nostrils, ear holes and other holes. In this way, the training and annotation data of multiple photos to the corresponding spatial point cloud are obtained.

[0054] The next step is to train the network.

[0055] The 3D AttriFlow model structure is as follows Figure 2 As shown, Figure 2 In the Attribute flow pipe in part (a), ResNet is a neural network used to convert the original image (one or more) into image representations. The AF module is an attribute flow module, which consists of two multi-layer neural networks that take the image representation output by ResNet and the point cloud position as input (see Figure 2 The output is divided into two parts: one is the geometric parameters σ and μ, which are the variance and mean of the AdaIn standardized output, respectively, obtained through a multi-layer neural network; the other is the s vector, which is the latent variable vector obtained after the image representation is reduced in dimension through a multi-layer neural network and then reduced in dimension through matrix decomposition. Figure 2 In the deformation pipe in part (b), the position of the initial point cloud (a set of spherical random points) is deformed, and the relative position relationship of the points is represented through several graph attention networks. The final point cloud is then output and compared with the groundtruth. The difference is calculated to obtain the loss function value.

[0056] In another exemplary embodiment of the present application, the vibration device includes:

[0057] A vibration motor, a shock absorbing layer and a control unit, wherein the shock absorbing layer wraps the vibration motor, the vibration motor is connected to the control unit, and the control unit is connected to the control module.

[0058] The vibration device of the present application adopts electromagnetic drive, and an eccentric wheel is driven by a motor to generate vibration. The frequency of vibration can be adjusted by changing the voltage of the motor. The principle is to generate a magnetic field by passing electricity around the electromagnet, and use the change of the magnetic field to drive the vibration plate to generate vibration, which is used to assist falling asleep and maintaining a deep sleep state. The vibration device can generate vibrations of different frequencies, and the frequency control is realized by a microcontroller.

[0059] In another exemplary embodiment of the present application, the sleep-aiding pillow further includes: a battery and a charging module, the battery and the charging module are connected, the battery is connected to a vibration device, the charging module supports USB or wireless charging, and has a built-in low-power design to extend battery life. The vibration device is powered by USB or wireless charging, so that the user can conveniently charge the pillow at any time. At the same time, the vibration device is low-power, and the durability and stability of long-term use are fully considered. It can not only ensure the long-term operation of the vibration device, but also avoid the interference of excessive power consumption to the user, so as to ensure that the power consumption can be kept within a reasonable range when it is running for a long time, so as to ensure the service life of the product.

[0060] In another exemplary embodiment of the present application, the sleep-aid pillow also includes: a pillowcase, made of environmentally friendly and biodegradable materials (organic cotton and bamboo fiber blended fabric, which is soft and breathable, has natural antibacterial and anti-odor functions, and is completely biodegradable, ensuring minimal impact on the environment when discarded), designed to be detachable and easy to clean to ensure that the user can enjoy a comfortable use experience while keeping the pillow clean and hygienic.

[0061] In another exemplary embodiment of the present application, in order to achieve the best use effect, the user can operate and monitor the pillow through a matching mobile application, transmit data via Bluetooth or Wi-Fi, and provide a user interface for monitoring sleep status and adjusting vibration frequency. The application can receive and process signals from the pillow, display the user's sleep data, and provide personalized suggestions to optimize the sleeping environment and habits. In addition, the application can also receive power information from the pillow and remind the user to charge when the power is low.

[0062] The mobile app provides personalized recommendations to optimize your sleep environment and habits by:

[0063] 1. The application analyzes the sleep data collected over a long period of time and identifies the user's sleep patterns, such as the time it takes to fall asleep, the number of times you wake up, the duration of deep sleep, etc.

[0064] 2. The application can be linked with other smart home devices, such as lights, thermostats and air purifiers, to monitor and adjust the user's sleeping environment (such as temperature, humidity, light, etc.).

[0065] 3. Recommend suitable sleeping and waking times based on the user’s actual sleep needs. For example, if the user falls asleep late, the system will remind the user to go to bed early and help the user adjust his or her sleep time.

[0066] 4. Recommendations for adjusting room temperature, light or humidity based on environmental monitoring data. For example, through the humidity sensor, it indicates that the humidity is too high and recommends dehumidification; or through the external sound and light sensor, it indicates that the ambient volume or brightness is too high to help users optimize their sleeping environment.

[0067] 5. By analyzing the behavior before going to bed, it is recommended to reduce the use of electronic devices, adjust eating habits, etc. to improve sleep quality. For example, when the proportion of deep sleep to total sleep is insufficient, it will prompt the user to take some measures to improve sleep to help the user improve his sleeping habits.

[0068] 6. The application generates a sleep quality report every week or month, listing in detail the factors that affect sleep (sleep time, sleep environment parameters (humidity, temperature, ambient volume, ambient brightness), sleep habits (percentage of deep sleep, number of tossing and turning)), and provides suggestions for further optimization based on user feedback.

[0069] The present application also takes into account the safety and durability of the product. In another exemplary embodiment of the present application, the material of the pillow body is environmentally friendly and biodegradable material that meets safety standards: a mixture of polylactic acid (PLA) and biodegradable polyurethane foam. This foam not only provides comfortable support, but also can be naturally degraded under appropriate conditions, reducing the burden on the environment.

[0070] The working process of the sleep-aiding pillow provided in the above embodiment is as follows:

[0071] The biosensor monitors the user's heart rate, breathing rate and activity level in real time. These data are transmitted to the microcontroller for processing. The microcontroller inputs the vibration frequency of the vibration device, the user's heart rate, breathing rate and activity level into the neural network model to obtain a control strategy, and controls the vibration frequency of the vibration device according to the control strategy.

[0072] The present application also provides an embodiment to describe the specific use process of the above-mentioned sleep-aiding pillow:

[0073] A user purchased this smart pillow and downloaded the accompanying mobile app by scanning the QR code. Under the guidance of the app, the user took pictures of his head and neck from the front, back, left and right and uploaded them to the app. The app automatically extracted the size and shape information of the user's head and neck through the built-in image processing algorithm and generated a 3D model.

[0074] Then, the 3D model is sent to a 3D printing factory and printed out by a high-precision 3D printer, forming a pillow body that fully conforms to the shape of the user's head and neck. The pillow body is made of environmentally friendly and biodegradable materials, which has good elasticity and support, and has good environmental performance.

[0075] The pillow itself is then sent to the assembly plant, where it is put into a specially designed pillowcase and is also equipped with a vibration device and biosensor. The vibration device can generate vibrations of different frequencies, while the biosensor can monitor and analyze the user's physiological status in real time, such as heart rate, breathing rate and activity level.

[0076] After assembly, the smart pillow was delivered to the user's home. The user activated the pillow through the mobile app and started using it. During use, the pillow monitors the user's physiological state in real time through biosensors and adjusts the vibration frequency of the vibration device based on the monitoring data through a microcontroller. At the same time, the app collects and analyzes data from the pillow, displays the user's sleep quality, and provides personalized sleep advice. When the pillow is low on power, the app will also remind the user to charge it in time.

[0077] This application provides a new sleep solution that combines advanced knowledge and technology in multiple fields. It not only provides personalized support, but also provides appropriate stimulation according to the user's sleep state, effectively improving the user's sleep quality. This design not only overcomes the limitations of existing products, but also opens up a new direction for the development of sleep technology. The advantages of this application are mainly reflected in the following aspects:

[0078] 1. This application innovatively combines knowledge and technologies in the fields of 3D printing technology, bioengineering, energy engineering and artificial intelligence, providing a new perspective and solution to solve human sleep problems. Compared with traditional sleep products, this application not only provides personalized physical support, but also provides appropriate stimulation according to the user's biological rhythm and sleep pattern, thereby more effectively improving sleep quality.

[0079] 2. The pillow material of this application is environmentally friendly and biodegradable. At the same time, the vibration device is low-power, and the control of electromagnetic radiation is also within the safety standard, which fully considers the safety and comfort of the user, the control of electromagnetic radiation and the management of power consumption. In this way, both in the selection of pillow materials and the design of the vibration device, a high safety standard is achieved.

[0080] 3. Compared with traditional sleep products, this application can more effectively improve sleep quality and cognitive function. This is because the use of 3D printing technology can ensure that the shape of the pillow fits the shape of the user's head and neck completely, allowing the pillow surface to fit the skin more closely, increasing the contact area, and providing support that is more in line with the user's needs, so that the vibration can be transmitted to the brain more efficiently, which can minimize the pressure on the head and neck, increase the comfort of sleep, and avoid chronic neck pain and shoulder tension caused by improper support for the head and neck, relieve cervical pain, and be more user-friendly for users with craniocerebral trauma or surgery, which is unmatched by traditional pillow products. The vibration device can generate vibrations of different frequencies according to the user's sleep state, helping the user to enter deep sleep faster and maintain a deep sleep state, and assisting in nerve repair in the process. The combination of the two makes the present application more effective than traditional sleep products in improving sleep quality, and has the new effect of improving cognitive function and preventing cognitive impairment.

[0081] 4. Through biosensors and microcontrollers, the present application can monitor and analyze the user's physiological state in real time, and then adjust the frequency of the vibration device according to the analysis results. This intelligent sleep aid method makes the effect of the present application far superior to traditional sleep aid products.

[0082] 5. Studies have shown that 40 Hz gamma waves can help people fall asleep faster and improve their cognitive functions, while 5 Hz slow-wave vibrations can help people fall asleep more deeply. However, this scientific theory is almost always applied to sound and light stimulation, which is often intolerable to users, thus reducing its sleep-aiding effect. This application controls the vibration frequency according to the user's heart rate, breathing rate, and activity level by setting up a vibration device and combining it with the optimal strategy provided by DQN. This design fully takes into account human biorhythms and brain activity patterns, and can provide timely stimulation according to the user's heart rate, breathing rate, and activity level, effectively improving the user's sleep quality.

[0083] 6. In terms of real-time personalized optimization for patients, this application uses a deep Q network (DQN) as the core reinforcement learning algorithm to adjust the output frequency of the vibration device in real time to achieve personalized adjustment of the user's circadian rhythm. The DQN neural network is used to continuously learn the user's physiological feedback (such as heart rate, body temperature, activity level, etc.) to optimize the adjustment strategy to help users enter and maintain a deep sleep state faster, thereby significantly improving sleep quality. The reinforcement learning algorithm has dynamic adaptability and can provide personalized adjustment plans based on the different physiological characteristics of each user.

[0084] 7. This application combines the deep Q network algorithm to generate vibration stimulation of the corresponding frequency at the appropriate time, effectively helping users enter and maintain deep sleep. This design is based on the working principle of frequency following reaction, which makes this application have a unique advantage in helping users enter and maintain deep sleep.

[0085] 8. This application introduces reinforcement learning algorithms, especially deep Q networks (DQN) as one of the core technologies, to achieve personalized circadian rhythm regulation for users. Through DQN, the system can learn and adjust according to the user's real-time physiological data (such as heart rate, body temperature, activity level, etc.), and automatically select the optimal vibration stimulation mode to promote high-quality sleep. DQN uses deep neural networks to process high-dimensional state space, and through continuous interaction and adaptive optimization, the system can provide personalized intervention measures based on the individual differences and physiological states of patients. This adaptive adjustment strategy based on reinforcement learning significantly improves the intelligence and regulation effect of the system.

[0086] 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, stored data, displayed data, 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 relevant data must comply with relevant regulations.

[0087] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 specification.

[0088] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A sleeping-aiding pillow, characterized in that: The sleeping-aiding pillow comprises: A pillow body, a biosensor, a microcontroller and a vibration device; the biosensor and the vibration device are both connected to the microcontroller; the microcontroller and the vibration device are arranged inside the pillow body; the biosensor is used to collect the user's heart rate, breathing frequency and activity level; the microcontroller is used to control the vibration frequency of the vibration device according to the vibration frequency of the vibration device, the user's heart rate, breathing frequency and activity level, so as to assist the user to fall asleep and maintain the user in a deep sleep state; the pillow body is printed using 3D printing technology based on a photo of the user's head and neck.

2. The sleeping-aiding pillow according to claim 1, characterized in that: The biosensor is an embedded biosensor.

3. The sleeping-aiding pillow according to claim 1, characterized in that: The microcontroller comprises: The control module is used to input the vibration frequency of the vibration device at the current moment, the user's heart rate, breathing rate and activity level into the frequency control model to obtain the regulation strategy of the vibration device, and control the vibration frequency of the vibration device at the current moment according to the regulation strategy; the regulation strategy is to increase the vibration frequency, reduce the vibration frequency or keep it unchanged; the frequency control model is trained by using a deep reinforcement learning algorithm.

4. The sleeping-aiding pillow according to claim 3, characterized in that: The deep reinforcement learning algorithm is the DQN algorithm.

5. The sleeping-aiding pillow according to claim 1, characterized in that: The pillow body construction process specifically includes: Obtaining a photograph of the user's head and neck; Use a deep learning model to perform face recognition on the user's head and neck photos to obtain the position of the user's head; The user's head and neck photo is processed using a contour model to obtain a 3D point cloud of the user's head and neck; the contour model is obtained by training the 3DAttriFlow model; Obtaining the size and shape of the user's neck based on the user's head and neck photo; Building a 3D model based on the position of the user's head, the 3D point cloud of the user's head and neck, and the size and shape of the user's neck; The 3D model is printed using 3D printing technology to obtain the pillow body.

6. The sleeping-aiding pillow according to claim 3, characterized in that: The vibration device comprises: A vibration motor, a shock absorbing layer and a control unit, wherein the shock absorbing layer wraps the vibration motor, the vibration motor is connected to the control unit, and the control unit is connected to the control module.

7. The sleeping-aiding pillow according to claim 1, characterized in that: The sleep-aiding pillow further comprises: a battery and a charging module, wherein the battery is connected to the charging module, and the battery is connected to the vibration device.

8. The sleeping-aiding pillow according to claim 1, characterized in that: The sleeping-aiding pillow further comprises a pillowcase.

Images