A sleep-aiding pillow based on artificial intelligence and physiological monitoring technology
By combining an external radar unit with an AI voice-guided sleep aid pillow and CBTI therapy, the problem of personalized intervention and monitoring accuracy of existing sleep aid pillows has been solved, achieving efficient and low-cost sleep management and improved user experience.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- ZHEJIANG YOUYOU MEDICAL TECHNOLOGY CO LTD
- Filing Date
- 2025-04-05
- Publication Date
- 2026-06-16
Smart Images
Figure CN224357876U_ABST
Abstract
Description
Technical Field
[0001] This utility model relates to the field of smart home and sleep health technology, specifically to a sleep aid pillow based on artificial intelligence and physiological monitoring technology. Background Technology
[0002] Insomnia is a common health problem worldwide. Traditional treatments for insomnia include medication and psychotherapy. However, medication may have side effects, while psychotherapy (such as cognitive behavioral therapy, CBT-I) requires guidance from a professional physician, is costly, and is difficult to popularize.
[0003] Traditional sleep aids often rely on material optimization (such as latex and memory foam) or simple functions (such as heating), but lack personalized intervention capabilities for insomniacs. In existing technologies, sleep monitoring devices and sleep aids are mostly independent modules, resulting in functional fragmentation, complex operation, and high costs. Furthermore, current sleep aids struggle to incorporate CBTI (Cognitive Behavioral Therapy) principles to achieve sleep time control and dynamic adjustment of intervention strategies, failing to meet users' demands for intelligent, integrated, and cost-effective sleep aids.
[0004] On the other hand, traditional sleep aids often use built-in sensors (such as pressure sensors and heart rate monitors), which result in bulky designs and difficulties in maintenance. In existing technologies, integrated sleep monitoring pillows increase the thickness of the pillow core due to the embedded radar module (affecting comfort), and require complete replacement when the hardware fails, leading to high costs. Furthermore, the built-in radar is susceptible to interference from pillow deformation, limiting monitoring accuracy.
[0005] Patent CN202411427066.9 discloses a processing system for sleep disorder intervention, comprising: a user terminal for collecting treatment period data information of patients with sleep disorders during cognitive behavioral therapy (CBTI) treatment; the treatment period data information includes patient data and treatment environment data; a sleep intervention analysis device for configuring the patient's CBTI treatment requirement information, wherein the CBTI treatment includes multiple treatment items, and the CBTI treatment requirement information includes the requirement information corresponding to each treatment item; and receiving the treatment period data information sent by the aforementioned user terminal, comparing the treatment period data information with the aforementioned CBTI treatment requirement information corresponding to the patient, and when it is determined that the patient's treatment period data information does not meet the CBTI treatment requirements, obtaining the difference information between the treatment period data information and the CBTI treatment requirement information, analyzing the aforementioned difference information to obtain the patient's treatment item gap information; and generating a sleep management suggestion report including the aforementioned treatment item gap information. The drawback of this solution is that it does not directly change the user's behavior through cognitive behavioral therapy (CBTI), but only provides suggestions, and does not combine the principles of CBTI (cognitive behavioral therapy) to achieve sleep time control and dynamically adjust intervention strategies.
[0006] In summary, existing sleep pillows have the following problems: the CBT-I theory is difficult to apply directly to sleep products, or requires operation or reminders from devices such as mobile phones, and the built-in radar and sensors are easily interfered with by pillow deformation, limiting the monitoring accuracy. Utility Model Content
[0007] The purpose of this invention is to provide a sleep aid pillow based on artificial intelligence and physiological monitoring technology to solve the problems of existing sleep pillows in terms of functional implementation and difficulty in dynamic adjustment.
[0008] To solve the above-mentioned technical problems, this application provides the following technical solution: A sleep aid pillow based on artificial intelligence and physiological monitoring technology, comprising a pillow body unit, a radar unit, a control unit, and a mobile unit; the pillow body unit includes a pillow core and a functional module layer; the radar unit is an external independent module connected to the pillow body unit via wireless communication; the control unit includes a remote controller and a main control box, with the remote controller being external; the mobile unit is used for sleep data analysis and the distribution of intervention strategies.
[0009] In one embodiment, the main control box is integrated into the bottom of the pillow core, and the main control box includes an air pump, a power management chip, and a data fusion chip.
[0010] In one embodiment, the functional module layer includes an airbag assembly, a temperature control assembly, an audio module, and an AI voice unit.
[0011] In one embodiment, the temperature control component operates at a temperature of 35-40℃, and achieves a temperature control accuracy of ±1℃ through a PID algorithm.
[0012] In one embodiment, the mobile unit includes a mobile app and a cloud server. The mobile app has a built-in sleep assessment model that generates a personalized CBTI plan based on questionnaire data and radar physiological indicators. The plan is synchronized to the pillow unit. Furthermore, after the plan is synchronized, some functions of the remote control are locked.
[0013] In one implementation, the radar unit collects the user's respiratory data and / or body movement data and transmits it to the main control box via Wi-Fi. The main control box synchronizes the data to a mobile APP. The APP's AI analyzes and generates intervention commands, which are then sent to the pillow unit to execute the inflation of the airbag assembly, or voice guidance or white noise playback, thereby automating the CBTI therapy.
[0014] In one implementation, the mobile app has a built-in sleep assessment algorithm, intervention program library, and data dashboard, while the cloud server stores user historical data and generates dynamic sleep optimization programs through AI.
[0015] In one implementation, the mobile unit can scan a code to bind / unbind the radar unit and the pillow, supporting multi-room, multi-device networking.
[0016] In one embodiment, the pillow core is made of latex or memory foam; the airbag assembly includes two horizontally distributed neck airbags connected to a built-in air pump via a gas transmission pipeline; the temperature control assembly is distributed on the surface of the pillow core and supports heating function; the AI voice unit includes a prompt module, a white noise module, a music module, a hypnosis module, and a customization module.
[0017] In one embodiment, the white noise module provides 40Hz white noise to aid sleep.
[0018] In one embodiment, the temperature control component and the airbag component are based on both user heating and neck stretching traction.
[0019] In one embodiment, the airbag assembly is cyclically inflated and deflated.
[0020] In one implementation, the prompting module prompts bedtime and wake-up time.
[0021] In one embodiment, the neck airbags are laterally distributed in the lower 1 / 3 of the pillow core, and the height of the sleep aid pillow after inflation is 10-11cm, with a deflation response time of ≤3 seconds.
[0022] In one embodiment, the remote control includes a Bluetooth module and a Wi-Fi module, and the surface of the remote control is provided with physical buttons and indicator lights. The physical buttons are used for power on / off, network connection, temperature adjustment, sound control, and inflation control.
[0023] In one embodiment, the radar unit includes an independent housing, a magnetic bracket, and a wireless communication module, and the radar has a detection range of 0.1-0.5 meters.
[0024] In one embodiment, the wireless communication module of the radar unit operates at a frequency of 60 GHz, has a detection range of 0.1-0.5 meters, and supports a sleep apnea warning function.
[0025] In one embodiment, the remote control has a network button, which, when pressed and held, triggers an indicator light to flash until a Bluetooth / Wi-Fi connection is successfully established.
[0026] In one embodiment, the radar unit detects that the user is awake or it is time to get up, and transmits the data to the main control box. The main control box then controls the AI voice unit to gently wake the user, reducing the time spent getting up. In another embodiment, the control module includes a stimulation control unit that detects the time the user has been in bed. If the user has not fallen asleep within a set time, the control unit guides the user to get out of bed via voice.
[0027] In one implementation, the radar unit is fixed to the headboard with a magnetic bracket, 20-30cm above the pillow; the radar scans the user's chest and abdomen at a frequency of 60GHz to generate a breathing waveform; when breathing pause is detected for more than 10 seconds, the main control box triggers the airbag to alternately inflate and deflate, guiding the user to adjust their sleeping position.
[0028] In one implementation, after the user gets out of bed, the radar unit continuously monitors the user's activity status; if the user does not return within 30 minutes, the AI voice unit pushes a notification via a mobile app: "We have detected that you have not fallen asleep yet, and we suggest you do relaxation training"; when the user returns to bed, the radar unit automatically reconnects to the pillow unit and resumes the personalized intervention process.
[0029] In one implementation, the user sets a target sleep schedule via the app; when the radar detects that the user has gone to bed early, a voice prompt is triggered saying "According to your sleep log, it is recommended to fall asleep later"; if the user does not fall asleep within 30 minutes, the airbag slowly deflates and plays a voice prompt to guide the user to get out of bed; when it is time to get up, the airbag inflates to raise the neck and plays progressive wake-up music simultaneously.
[0030] In one implementation, the mobile app generates an initial plan based on user questionnaire data, and the plan is sent to the pillow unit. The radar monitors the breathing rate in real time and dynamically adjusts the white noise intensity and voice guidance rhythm. If breathing apnea is detected, the airbag assembly is activated to inflate and adjust the sleeping position.
[0031] In one implementation, the user completes an insomnia assessment questionnaire via a mobile app, which generates an initial plan: target bedtime of 23:00 and wake-up time of 7:00. When the radar unit detects that the user goes to bed early at 22:30, the pillow's AI voice prompts: "Detected that you have not yet reached your planned bedtime, we suggest reading to relax." If the user remains awake for more than 25 minutes, the app simultaneously pushes a cognitive training course to the phone, while the airbag deflates to reduce neck support, prompting the user to get out of bed.
[0032] In one implementation, the radar unit detects that the number of nighttime apneas exceeds the limit and uploads the data to the cloud in real time; the mobile APP analyzes the data with AI and suggests the following day: "We have detected that you have mild sleep apnea. We have turned off the neck airbag and added side-sleeping guidance voice." The user can restore the original plan with one click via remote control, and the feedback data will be used for AI model iteration.
[0033] Compared with the prior art, the advantages of this utility model are:
[0034] Unlike existing technologies, this application is the first to deeply integrate cognitive behavioral therapy (CBTI) with hardware devices. It uses radar monitoring data to drive airbag inflation and deflation, voice guidance, and white noise adjustment in real time, realizing a closed loop of "monitoring-analysis-intervention" to directly change the user's sleep behavior (such as forced bed exit and timed wake-up), rather than just providing suggestions. Furthermore, the external radar unit is independently deployed through a magnetic bracket, avoiding pillow deformation interference with the signal, increasing the apnea detection rate to 92% (compared to only 78% for traditional built-in solutions), while reducing hardware maintenance costs by more than 60%.
[0035] Unlike existing technologies, this application adopts a multimodal interaction system: it supports remote management via mobile APP (long-term strategy), local control via remote control (emergency operation), and real-time AI voice guidance (dynamic intervention) to meet the needs of different scenarios; in addition, it also has dynamic anti-interference capabilities: it adopts dual-frequency communication (2.4GHz / 5GHz adaptive switching) and filtering algorithms to ensure stable operation in multi-device environments (such as mobile phones and air conditioners).
[0036] Unlike existing technologies, this application adopts a modular architecture, allowing the pillow body, radar, and controller to be produced and replaced independently, resulting in a significant improvement in yield. Furthermore, it supports global multi-language version switching via APP scanning and binding, enabling market expansion without hardware adjustments and demonstrating rapid adaptability.
[0037] Unlike existing technologies, this application can also wake up users with gentle sounds, vibrations, or heating functions based on their sleep depth and set wake-up time, and combine AI algorithms to analyze sleep data and optimize sleep plans.
[0038] Unlike existing technologies, this application also features a multi-user mode and simulation interaction, supporting independent use by multiple users. Through simulation interaction, it guides users to establish correct sleep cognition, enhances user experience, and can even convey messages between different users through the pillow, making it highly interactive.
[0039] Unlike existing technologies, based on the principles of CBTI (Behavioral Cognitive Therapy), the sleep-aid pillow of this application is suitable for use by people with insomnia. This pillow has an AI voice sleep-aid assistant, muscle relaxation and soothing, sleep-aid white noise, and intelligent sleep cognitive control functions (sleep time control). At the same time, the pillow unit is connected to a smart APP. The smart APP can assess the user's situation and send the assessment results to the AI module of the sleep-aid pillow. The AI module will then develop a personalized sleep-aid plan, and the radar sleep monitoring device will provide the sleep monitoring results to the APP.
[0040] The embodiments of this application can achieve other advantageous technical effects not listed one by one. These other technical effects may be partially described below and can be expected and understood by those skilled in the art after reading this application. Attached Figure Description
[0041] Figure 1 This is a schematic diagram of the overall structure of the sleep-aid pillow of this utility model.
[0042] Figure 2 This diagram shows the structure of the pillow body unit and the two main modules of the control unit of this utility model.
[0043] Figure 3 This is a flowchart illustrating the sleep-aid pillow of this utility model.
[0044] The features represented by the numbers in the attached diagram are as follows:
[0045] 1-Pillow body unit, 11-Pillow core, 12-Functional module layer, 2-Radar unit, 3-Control unit, 31-Remote controller, 32-Main control box, 4-Moving unit. Detailed Implementation
[0046] To make the objectives, features, and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0047] The present application will now be described in more detail with reference to various embodiments and examples of several aspects thereof. Specific Implementation Example 1
[0048] This embodiment provides a solution to the aforementioned technical problems, achieved through the following technical solution: a sleep aid pillow based on artificial intelligence and physiological monitoring technology, comprising a pillow body unit 1, a radar unit 2, a control unit 3, and a moving unit 4; the pillow body unit 1 includes a pillow core 11 and a functional module layer 12; the radar unit 2 is an external independent module connected to the pillow body unit 1 via wireless communication; the control unit 3 includes a remote controller 31 and a main control box 32, the remote controller 31 being external; the moving unit 4 is used for sleep data analysis and intervention strategy distribution, such as... Figure 1 and Figure 2 As shown.
[0049] In this first embodiment, the main control box 32 is integrated into the bottom of the pillow core 11. The main control box 32 includes an air pump, a power management chip, and a data fusion chip.
[0050] In this first embodiment, the functional module layer 12 includes an airbag assembly, a temperature control assembly, an audio module, and an AI voice unit.
[0051] In this first embodiment, the operating temperature of the temperature control component is 35-40℃, and the temperature control accuracy is ±1℃ through the PID algorithm.
[0052] In this first embodiment, the mobile unit 4 includes a mobile app and a cloud server. The mobile app has a built-in sleep assessment model that generates a personalized CBTI plan through questionnaire data and radar physiological indicators. The plan is synchronized to the pillow unit 1. After the plan is synchronized, some functions of the remote control 31 will be locked.
[0053] In this first embodiment, the radar unit 2 collects the user's respiratory data and / or body movement data and transmits it to the main control box 32 via Wi-Fi. The main control box 32 synchronizes the data to the mobile APP. The AI on the APP analyzes and generates intervention commands, which are then sent to the pillow body unit 1 to execute the inflation of the airbag assembly, or voice guidance or white noise playback, thereby automating the CBTI therapy.
[0054] In this first embodiment, the mobile app has a built-in sleep assessment algorithm, intervention plan library and data dashboard, and the cloud server stores user historical data and generates dynamic sleep optimization plans through AI.
[0055] In this first embodiment, the mobile unit 4 can scan a code to bind / unbind the radar unit 2 and the pillow body, supporting multi-room and multi-device networking.
[0056] In this embodiment, the pillow core 11 is made of latex or memory foam; the airbag assembly includes two horizontally distributed neck airbags connected to a built-in air pump via a gas transmission pipeline; the temperature control assembly is distributed on the surface of the pillow core 11 and supports heating function; the AI voice unit includes a prompt module, a white noise module, a music module, a hypnosis module, and a customization module.
[0057] In this first embodiment, the white noise module provides 40Hz white noise to aid sleep.
[0058] In this first embodiment, the temperature control component and the airbag component are based on both user heating and neck stretching traction.
[0059] In this first embodiment, the airbag assembly is cyclically inflated and deflated.
[0060] In this first embodiment, the prompting module prompts the bedtime and wake-up time.
[0061] In this embodiment, the neck airbags are distributed laterally in the lower 1 / 3 of the pillow core 11. After inflation, the height of the sleep aid pillow is 10-11cm, and the deflation response time is ≤3 seconds.
[0062] In this first embodiment, the remote control 31 includes a Bluetooth module and a Wi-Fi module, and the surface of the remote control 31 is provided with physical buttons and indicator lights. The physical buttons are used for power on / off, network connection, temperature adjustment, sound control, and inflation control.
[0063] In this first embodiment, the radar unit 2 includes an independent housing, a magnetic bracket, and a wireless communication module, and the radar has a detection range of 0.1-0.5 meters.
[0064] In this first embodiment, the wireless communication module of the radar unit 2 operates at a frequency of 60GHz, has a detection range of 0.1-0.5 meters, and supports the function of sleep apnea warning.
[0065] In this first embodiment, the remote control 31 is equipped with a network button. After pressing and holding the button, the indicator light will flash until the Bluetooth / Wi-Fi connection is successful.
[0066] In this first embodiment, the radar unit 2 detects that the user is awake or it is time to get up, and transmits the data to the main control box 32. The main control box 32 then controls the AI voice unit to gently wake the user, reducing the time spent getting up. In this first embodiment, the control module includes a stimulation control unit 3. The stimulation control unit 3 detects the time the user has been in bed. If the user has not fallen asleep within the set time, the system guides the user to get out of bed via voice.
[0067] In this first embodiment, the radar unit 2 is fixed to the head of the bed with a magnetic bracket, 20-30cm above the pillow surface; the radar scans the user's chest and abdomen with a 60GHz frequency band and generates a breathing waveform; when breathing pause is detected for more than 10 seconds, the main control box 32 triggers the airbag to alternately inflate and deflate, guiding the user to adjust their sleeping position.
[0068] In this first embodiment, after the user gets out of bed, the radar unit 2 continuously monitors the user's activity status; if the user does not return within 30 minutes, the AI voice unit pushes a prompt through the mobile APP: "It has been detected that you have not fallen asleep yet. We suggest you do relaxation training." When the user returns to bed, the radar unit 2 automatically reconnects to the pillow unit 1 and resumes the personalized intervention process.
[0069] In this first embodiment, the user sets a target sleep schedule through the APP; when the radar detects that the user has gone to bed early, it triggers a voice prompt "According to your sleep log, it is recommended to fall asleep later"; if the user does not fall asleep within 30 minutes, the airbag slowly deflates and plays a voice prompt to guide the user to get out of bed; when it is time to get up, the airbag inflates to raise the neck and plays a progressive wake-up music at the same time.
[0070] In this first embodiment, the mobile app generates an initial plan based on user questionnaire data, and the plan is sent to the pillow unit 1. The radar monitors the breathing rate in real time and dynamically adjusts the white noise intensity and voice guidance rhythm. If breathing apnea is detected, the airbag assembly is activated to inflate and adjust the sleeping position.
[0071] In this first embodiment, the user completes an insomnia assessment questionnaire via a mobile app, which generates an initial plan: target bedtime 23:00 and wake-up time 7:00. The radar unit 2 detects that the user goes to bed early at 22:30, and the pillow's AI voice prompts: "Detected that you have not yet reached your planned bedtime, we suggest reading to relax." If the user remains awake for more than 25 minutes, the app simultaneously pushes a cognitive training course to the phone, while the airbag deflates to reduce neck support, prompting the user to get out of bed.
[0072] In this first embodiment, the radar unit 2 detects that the number of nighttime apneas by the user exceeds the standard and uploads the data to the cloud in real time; the mobile APP analyzes the data through AI and suggests the following day: "We have detected that you have mild sleep apnea. We have turned off the neck airbag for you and added side sleeping guidance voice". The user can restore the original plan with one click through the remote control 31. The feedback data will be used for AI model iteration.
[0073] In this embodiment, as Figure 3 As shown, the operating process of the sleep pillow is as follows:
[0074] Initial setup: Users fill out an insomnia assessment questionnaire (such as age, cause of insomnia, and sleep goals) via mobile app or voice; AI generates a personalized CBTI plan (such as target sleep schedule of 23:00-7:00).
[0075] Sleep monitoring phase: After the user lies down, radar unit 2 continuously monitors the user's breathing rate, body movement data and sleep stage (awake / light sleep / deep sleep), with a detection distance of 0.1-0.5m and an accuracy of ±1 breath / minute; the data is transmitted to the main control box 32 via WiFi and synchronized to the mobile APP and cloud server.
[0076] Dynamic intervention execution phase: If the user goes to bed early (e.g., 10:30 PM), the AI voice prompts "It is recommended to delay going to sleep." If the user remains awake for more than 25 minutes, the airbag deflates to lower the support height, and the APP pushes suggestions for getting out of bed (e.g., meditation training). When it is time to get up (7:00 AM), the airbag inflates to raise the neck, plays progressive wake-up music, and forcibly locks the "delayed wake-up" function on the remote control 31 to ensure a regular sleep schedule. During sleep, the intensity of 40Hz white noise is dynamically adjusted according to the sleep stage, the PID algorithm maintains the pillow surface temperature at 35-40℃±1℃, the neck airbag is periodically inflated and deflated, and the AI voice module calls up hypnotic scripts such as progressive relaxation and the stair-climbing method according to the plan. The voice tone supports switching between multiple roles.
[0077] Data analysis and optimization: The APP generates a sleep efficiency report (actual sleep / bedtime), and radar data marks abnormal breathing events; cloud AI analyzes historical data to optimize the next cycle of intervention plan (such as reducing the allowed bedtime to 6 hours)).
[0078] The foregoing description of several embodiments of this application has been provided for illustrative purposes. This foregoing description is not intended to be exhaustive, nor is it intended to limit this application to the precise configurations, structures, and / or steps disclosed. Obviously, based on the foregoing teachings, those skilled in the art can make various modifications and improvements without departing from the inventive concept, and these all fall within the protection scope of this utility model. Therefore, the protection scope of this utility model patent should be determined by the appended claims.
Claims
1. A sleep-aid pillow based on artificial intelligence and physiological monitoring technology, characterized in that, It includes a pillow body unit, a radar unit, a control unit, and a mobile unit; the pillow body unit includes a pillow core and a functional module layer; the radar unit is an external independent module that is connected to the pillow body unit via wireless communication; the control unit includes a remote controller and a main control box, with the remote controller being external; the mobile unit is used for sleep data analysis and the distribution of intervention strategies.
2. The sleep-aid pillow according to claim 1, characterized in that, The main control box is integrated at the bottom of the pillow core, and the main control box includes an air pump, a power management chip, and a data fusion chip.
3. The sleep-aid pillow according to claim 1, characterized in that, The functional module layer includes an airbag assembly, a temperature control assembly, an audio module, and an AI voice unit.
4. The sleep-aid pillow according to claim 1, characterized in that, The mobile unit includes a mobile app and a cloud server. The mobile app has a built-in sleep assessment model that generates a personalized CBTI plan based on questionnaire data and radar physiological indicators. The plan is synchronized to the pillow unit. Furthermore, after the plan is synchronized, some functions of the remote control are locked.
5. The sleep-aid pillow according to claim 3, characterized in that, The radar unit collects the user's respiratory data and / or body movement data and transmits it to the main control box via Wi-Fi. The main control box synchronizes the data to the mobile APP. The AI on the APP analyzes and generates intervention commands, which are then sent to the pillow unit to execute the inflation of the airbag assembly, or voice guidance or white noise playback, thereby automating the CBTI therapy.
6. The sleep-aid pillow according to claim 3, characterized in that, The pillow core is made of latex or memory foam; the airbag assembly includes two horizontally distributed neck airbags connected to a built-in air pump via a gas transmission pipeline; the temperature control assembly is distributed on the surface of the pillow core and supports heating function; the AI voice unit includes a prompt module, a white noise module, a music module, a hypnosis module, and a customization module.
7. The sleep-aid pillow according to claim 1, characterized in that, The remote control includes a Bluetooth module and a Wi-Fi module, and the surface of the remote control is provided with physical buttons and indicator lights. The physical buttons are used for power on / off, network connection, temperature adjustment, sound control, and inflation control.
8. The sleep-aid pillow according to claim 1, characterized in that, The radar unit includes an independent housing, a magnetic bracket, and a wireless communication module. The radar has a detection range of 0.1-0.5 meters.
9. The sleep-aid pillow according to claim 1, characterized in that, The remote control has a network button. Pressing and holding the button will trigger an indicator light to flash until the Bluetooth / Wi-Fi connection is successful.
10. The sleep-aid pillow according to claim 3, characterized in that, The radar unit detects that the user is awake or it is time to get up, and transmits the data to the main control box. The main control box then controls the AI voice unit to gently wake the user, reducing the time spent getting up.