Old people health monitoring robot system based on cloud intelligent control in pension institution

Abstract

The invention relates to the technical field of medical health, in particular to an old people health monitoring robot system for an old people institution based on cloud intelligent control. The system comprises a health monitoring management generation module, a health monitoring execution acquisition module, a health activity abnormity analysis module and a remote monitoring alarm module. Historical physiological health data and historical daily activity data can be acquired, and a health monitoring management scheme is generated by using a cloud control platform. A cloud control platform is utilized to send a health monitoring task instruction to a corresponding health monitoring robot and execute navigation to the position of the old in an old-age care institution, meanwhile, medicine delivery and meal delivery tasks are executed, body recognition inquiry and physiological health signal collection are carried out, and meanwhile, health signal abnormity positioning and early warning response are carried out. And generating a health abnormity early warning signal corresponding to the elderly in the pension institution, and timely sending early warning information to medical personnel or family members corresponding to the pension institution. The health monitoring management efficiency of the old-age care institution can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical and health, and particularly to an elderly health monitoring robot system for pension institutions based on cloud intelligent control. Background Art

[0002] With the intensification of population aging, the health monitoring and night safety guarantee of the elderly in pension institutions have become social problems to be solved urgently. At present, with the progress of technology, intelligent technology has begun to be gradually applied to the pension industry, and many pension institutions have begun to introduce health monitoring devices, such as smart bracelets, smart mattresses, etc., to monitor the physical conditions of the elderly. However, most of these devices work independently, and there are the following technical pain points: 1. Insufficient night monitoring manpower: The number of night shift personnel in pension institutions is limited, and it is difficult to monitor all the elderly in real time; 2. Scattered monitoring devices: The existing monitoring devices are mainly fixed cameras, with monitoring blind spots and lacking the ability of active patrol inspection; 3. Lag in emergency response: When emergencies such as the elderly falling or having difficulty breathing occur, it is difficult for fixed devices to detect and handle them in time; 4. Low degree of intelligence: The existing devices lack intelligent analysis and early warning capabilities, and cannot achieve active monitoring and abnormal early warning of the physiological state of the elderly; 5. Serious information silos: There is a lack of data interconnection and collaboration between various monitoring devices, and it is difficult to form a complete health monitoring closed loop. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide an elderly health monitoring robot system for pension institutions based on cloud intelligent control to solve at least one of the above technical problems.

[0004] To achieve the above object, an elderly health monitoring robot system for pension institutions based on cloud intelligent control includes the following modules: A health monitoring management generation module, configured to obtain the historical physiological health data and historical daily activity data corresponding to the elderly in the pension institution, and generate a health monitoring management plan corresponding to the elderly in the pension institution based on the historical physiological health data and historical daily activity data and by using a cloud control platform; The health monitoring execution collection module is used to send corresponding health monitoring task instructions to the corresponding health monitoring robot based on the health monitoring management plan of the elderly in the elderly care institution and using the cloud control platform, and based on the health monitoring task instructions, control the corresponding navigation and autonomous movement system of the health monitoring robot to navigate to the location of the elderly in the elderly care institution. At the same time, it acts on the corresponding environmental interaction system to perform medicine delivery and meal delivery tasks according to the predetermined health monitoring tasks in the health monitoring management plan, and uses the elderly health information collection system to conduct body recognition inquiries and physiological health signal collection to obtain the corresponding physical health feelings and physiological health signal spectrograms of the elderly in the elderly care institution; The health activity abnormality analysis module is used to perform health signal abnormality localization on the physiological health signal spectrogram of the elderly in the elderly care institution based on the corresponding physical health feelings of the elderly in the elderly care institution to obtain the corresponding health signal abnormal sites of the elderly in the elderly care institution; The activity tracking posture status of the elderly in the elderly care institution is monitored in real time through the health monitoring robot, and activity posture abnormality analysis is performed according to the corresponding activity tracking posture status of the elderly in the elderly care institution to obtain the corresponding activity tracking posture abnormal points of the elderly in the elderly care institution; The remote monitoring and alarm module is used to perform early warning response according to the corresponding health signal abnormal sites or activity tracking posture abnormal points of the elderly in the elderly care institution and using the cloud control platform, generate the corresponding health abnormality early warning signal of the elderly in the elderly care institution, and timely send warning information to the corresponding medical staff or family members of the elderly care institution.

[0005] Furthermore, the health monitoring management generation module includes the following functions: Obtain the corresponding historical physiological health data of the elderly in the elderly care institution, including heart rate, respiratory rate, blood pressure and body temperature; Obtain the corresponding historical daily activity data of the elderly in the elderly care institution, including daily activity volume, movement trajectory and posture changes; Transmit the corresponding historical physiological health data and historical daily activity data of the elderly in the elderly care institution into the corresponding cloud control platform, and use the cloud control platform to perform preliminary preprocessing on the historical physiological health data and historical daily activity data, including denoising, filtering and standardization, to remove the corresponding noise and interference information in the data. At the same time, the preprocessed historical physiological health data and historical daily activity data are classified and labeled for storage to generate the corresponding historical health monitoring data file of the elderly in the elderly care institution; Use the cloud control platform and combine the corresponding historical health monitoring data file of the elderly in the elderly care institution to generate the corresponding health monitoring management plan for the elderly in the elderly care institution in response to the health monitoring robot.

[0006] Further, the health monitoring and management plan corresponding to the elderly in the elderly care institution specifically includes corresponding health monitoring autonomous navigation, health medication dosage, healthy diet status, and activity tracking and monitoring management tasks.

[0007] Further, the health monitoring execution and acquisition module includes the following functions: Based on the health monitoring and management plan corresponding to the elderly in the elderly care institution and using the cloud control platform to send the health monitoring task instructions corresponding to the health monitoring and management plan to the corresponding health monitoring robot, and according to the local control unit corresponding to the health monitoring robot receiving and responding to the corresponding health monitoring task instructions to execute the health monitoring tasks corresponding to the elderly in the elderly care institution; Based on the health monitoring task instructions acting on the corresponding local control unit to respond and control the navigation and autonomous movement system of the health monitoring robot to execute autonomous navigation to the location of the elderly in the elderly care institution; Based on the health monitoring task instructions acting on the environmental interaction system of the health monitoring robot to execute the medicine delivery and meal delivery tasks corresponding to the elderly in the elderly care institution according to the pre-determined health medication dosage and healthy diet status management tasks in the health monitoring and management plan, to autonomously navigate to the location of the elderly in the elderly care institution according to the medicine delivery and meal delivery tasks, and to realize the door opening and closing operation of the corresponding room at the location through the degrees of freedom robotic arm and the anti-slip gripper, and at the same time use the degrees of freedom robotic arm and the anti-slip gripper to safely and accurately complete the transfer of the health medicines and dining items corresponding to the elderly in the elderly care institution; Use the elderly health information acquisition system of the health monitoring robot to conduct body recognition inquiries and physiological health signal acquisitions to obtain the physical health feelings and physiological health signal spectrograms corresponding to the elderly in the elderly care institution.

[0008] Further, the above-mentioned based on the health monitoring task instructions acting on the corresponding local control unit to respond and control the navigation and autonomous movement system of the health monitoring robot to execute autonomous navigation to the location of the elderly in the elderly care institution includes: Based on the health monitoring task instructions acting on the corresponding local control unit to respond and control the navigation and autonomous movement system of the health monitoring robot to construct a three-dimensional map of the monitoring environment corresponding to the elderly care institution to accurately locate the specific location of the elderly to be monitored in the elderly care institution; Based on the three-dimensional map of the monitoring environment and combined with The algorithm predicts the distribution of static obstacles between the location of the elderly in the elderly care institution and the health monitoring robot, and based on the distribution of static obstacles, conducts path planning for the health monitoring robot to generate the running distribution trajectory between the location of the elderly and the health monitoring robot; Obtain the current corresponding motion position, motion speed, motion direction, and the distribution of real-time environmental obstacles around the health monitoring robot based on the operation distribution trajectory between the location of the elderly and the health monitoring robot, and analyze the feasible motion window of the health monitoring robot at each moment based on the current corresponding motion position, motion speed, motion direction, and the distribution of real-time environmental obstacles around the health monitoring robot on the trajectory, and combine with the dynamic window method; Optimize the target trajectory for the feasible motion window of the health monitoring robot at each moment based on the motion direction and the distribution of real-time environmental obstacles around, so as to determine the turning corner deviation between the health monitoring robot and the distribution of each surrounding real-time environmental obstacle in the motion window according to the motion direction, and plan and optimize the obstacle avoidance running trajectory corresponding to the health monitoring robot in the feasible motion window at each moment based on the corresponding turning corner deviation, so as to generate the target optimized motion trajectory corresponding to the health monitoring robot in each motion window; Autonomously navigate the corresponding health monitoring robot to the location of the elderly in the elderly care institution based on the target optimized motion trajectory corresponding to the health monitoring robot in each motion window.

[0009] Further, the determination of the turning corner deviation between the health monitoring robot and the distribution of each surrounding real-time environmental obstacle in the motion window according to the motion direction includes: Establish a local coordinate system with the current position of the health monitoring robot as the origin in the motion window, and determine the corresponding direction position of the health monitoring robot and the distribution positions of each surrounding real-time environmental obstacle in the motion window based on the motion direction according to the local coordinate system; Conduct relative position angle analysis based on the corresponding direction position of the health monitoring robot and the distribution positions of each surrounding real-time environmental obstacle, so as to calculate the relative position angles between the health monitoring robot and each surrounding real-time environmental obstacle using geometric algorithms; Obtain the maximum acceleration and maximum turning rate corresponding to the health monitoring robot in the motion window, and calculate the turning corner deviation for the relative position angles between the health monitoring robot and each surrounding real-time environmental obstacle based on the maximum acceleration and maximum turning rate, so as to obtain the turning corner deviation between the health monitoring robot and the distribution of each surrounding real-time environmental obstacle in the motion window.

[0010] Further, the body recognition inquiry and physiological health signal collection using the elderly health information collection system corresponding to the health monitoring robot include: Use the elderly health information collection system corresponding to the health monitoring robot to conduct body recognition inquiries, so as to actively inquire about the physical feelings of the elderly in the pension institution during the medicine delivery and meal delivery processes corresponding to the health monitoring robot through speech recognition, and monitor and identify the call for help and groan feeling information of the elderly in the pension institution during this period, so as to obtain the physical health feelings of the elderly in the pension institution; Use the physiological health sign monitoring in the elderly health information collection system corresponding to the health monitoring robot to monitor the physiological health status, so as to collect the physiological health signals of the heart rate, respiratory rate and body temperature of the elderly in the pension institution through the non-contact sensors corresponding in the health monitoring robot, and obtain the physiological health signal status of the elderly in the pension institution; Perform signal spectrum conversion according to the physiological health signal status of the elderly in the pension institution to obtain the physiological health signal spectrogram of the elderly in the pension institution.

[0011] Furthermore, the health activity abnormality analysis module includes the following functions: Perform health signal abnormality positioning on the physiological health signal spectrogram of the elderly in the pension institution based on the physical health feelings of the elderly in the pension institution to obtain the health signal abnormal sites of the elderly in the pension institution; Real-time monitor the activity tracking posture status of the elderly in the pension institution through the health monitoring robot, including the daily activity volume and sleep body movement posture of the elderly in the pension institution; Perform time series change analysis on the activity tracking posture status of the elderly in the pension institution to generate the activity posture time change sequence of the elderly in the pension institution; Perform activity posture abnormality analysis according to the activity posture time change sequence of the elderly in the pension institution. If it is for the time change sequence corresponding to the daily activity volume, then statistically analyze the sudden activity fall time point of the elderly in the pension institution according to the time change sequence corresponding to the daily activity volume; if it is for the time change sequence corresponding to the sleep body movement posture, then determine the corresponding sleep body movement frequency according to the time change sequence corresponding to the sleep body movement posture, and based on the sleep body movement frequency, determine the sleep activity abnormal nodes for the time change sequence corresponding to the sleep body movement posture to determine the corresponding sleep activity abnormal time nodes, and merge the sudden activity fall time point and the sleep activity abnormal time nodes of the elderly in the pension institution to obtain the activity tracking posture abnormal points of the elderly in the pension institution.

[0012] Furthermore, the performing health signal abnormality positioning on the physiological health signal spectrogram of the elderly in the pension institution based on the physical health feelings of the elderly in the pension institution includes: Determine the corresponding call for help and groan time periods according to the call for help and groan feeling information in the physical health feelings of the elderly in the pension institution; Locate the abnormal health signal points on the physiological health signal spectrogram corresponding to the elderly in the elderly care institution based on the call-for-help and groaning time periods of the elderly in the elderly care institution, so as to determine the signal abnormal time points on the physiological health signal spectrogram during this time period according to the call-for-help and groaning time periods, including the rapid rise, rapid fall, and irregular oscillation time points, in order to obtain the abnormal health signal points corresponding to the elderly in the elderly care institution.

[0013] Furthermore, the remote monitoring and alarm module includes the following functions: Generate the corresponding health abnormality feedback information or activity posture abnormality feedback information for the elderly in the elderly care institution based on the abnormal health signal points or activity tracking posture abnormal points corresponding to the elderly in the elderly care institution, and feedback and upload it to the cloud control platform; Use the cloud control platform to respond to the early warning response according to the corresponding health abnormality feedback information or activity posture abnormality feedback information of the elderly in the elderly care institution, generate the corresponding health abnormality early warning signal for the elderly in the elderly care institution, and timely send the corresponding early warning information of the elderly in the elderly care institution to the medical staff or family members corresponding to the elderly care institution.

[0014] Advantages of the present invention: The elderly health monitoring robot system based on cloud intelligent control proposed by the present invention is generally composed of a health monitoring management generation module, a health monitoring execution and acquisition module, a health activity anomaly analysis module, and a remote monitoring and alarm module. Compared with the prior art, the beneficial effect of this application lies in that obtaining the historical physiological health data and daily activity data of the elderly in the elderly care institution is the key to constructing a health monitoring management plan. By collecting and analyzing these data, the health status and living habits of the elderly can be comprehensively understood, and a personalized health management file can be formed. The key to this process is that it can solve the problem of information islands in traditional elderly care institutions and break the situation where various health data and activity data are scattered. Through the centralized management of the cloud control platform, the elderly care institution can track and analyze the health status of each elderly person in real time, making the management more accurate and comprehensive. This not only improves the intelligent level of elderly care services but also can customize a personalized health management plan for each elderly person, including corresponding health monitoring autonomous navigation, health medication dosage, healthy diet status, and activity tracking and monitoring management tasks, to anticipate potential health problems in advance and further improve the resource utilization rate of the elderly care institution. Secondly, after receiving accurate health monitoring task instructions through the cloud platform, the health monitoring robot can efficiently execute health management tasks. The key to this process is that it can solve the problems of insufficient manpower, scattered equipment, and lagging emergency response in traditional elderly care institutions. The health monitoring robot can not only automatically navigate to the location of the elderly person to avoid the lag in monitoring caused by manpower shortage but also independently execute daily tasks such as delivering medicine and meals to ensure that the elderly person receives timely life care and nursing. This automated operation reduces the need for manpower, enabling nursing staff to concentrate their energy on tasks that require more interpersonal care and improving work efficiency. In addition, the combination of the elderly health information acquisition system enables the robot to obtain the physiological health signals of the elderly in real time and generate a physiological health signal spectrogram based on these data. By continuously monitoring the health status of the elderly, the system can timely adjust the nursing strategy and provide customized health management for the elderly. Then, by using the physiological health signal spectrogram for health signal anomaly localization, the health anomalies of the elderly can be effectively identified. The key to this process is that it can real-time detect potential problems in health signals, such as abnormal heart rate and blood pressure fluctuations, to help medical staff intervene early. This precise health monitoring method can reduce the risk of sudden illness or health crisis and enhance the safety guarantee of the elderly. At the same time, the health monitoring robot can also monitor the activities of the elderly in real time, track and analyze the activity postures of the elderly, and identify abnormal actions or behaviors. This activity tracking and abnormal posture analysis helps to detect corresponding falls or abnormal sleep body movements during the movement of the elderly, issue early warnings, and take countermeasures.Finally, through the cloud control platform, automatic early warning of health abnormalities is realized, significantly enhancing the emergency response ability of pension institutions. When health signals or activity postures are abnormal, warning signals can be quickly sent out and information can be transmitted to relevant medical staff or family members. This early warning mechanism not only improves the efficiency of handling emergencies but also effectively avoids the deterioration of health problems caused by lagging emergency responses. Especially at night or when unattended, the intelligent system can monitor the health status of the elderly around the clock to ensure that any abnormal situation can be dealt with in a timely manner, greatly enhancing the sense of security of the elderly and the trust of their family members. By reasonably arranging the work of nursing staff, it ensures that each elderly person can receive professional health care at the most appropriate time. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments when read in conjunction with the accompanying drawings: Figure 1 It is a schematic diagram of the modules of the elderly health monitoring robot system in a pension institution based on cloud intelligent control of the present invention; Figure 2 For Figure 1 it is a schematic diagram of the functional flow of the health monitoring management generation module in Figure 3 For Figure 1 it is a schematic diagram of the functional flow of the health monitoring execution and acquisition module in DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following clearly and completely describes the technical system of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those skilled in the art within the scope of the present invention without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0017] In addition, the accompanying drawings are only schematic diagrams of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor systems and / or microcontroller systems.

[0018] It should be understood that although terms such as "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly, the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0019] To achieve the above object, please refer to Figures 1 to 3 , the present invention provides a health monitoring robot system for the elderly in a pension institution based on cloud intelligent control. The system includes the following modules: A health monitoring management generation module, configured to obtain the historical physiological health data and historical daily activity data corresponding to the elderly in the pension institution, and generate a health monitoring management plan corresponding to the elderly in the pension institution based on the historical physiological health data, historical daily activity data, and using a cloud control platform; A health monitoring execution and collection module, configured to send corresponding health monitoring task instructions to the corresponding health monitoring robot based on the health monitoring management plan corresponding to the elderly in the pension institution and using the cloud control platform, and based on the health monitoring task instructions, control the corresponding navigation and autonomous movement system of the health monitoring robot to navigate to the location of the elderly in the pension institution, and at the same time act on the corresponding environment interaction system to perform tasks such as delivering medicine and meals according to the health monitoring tasks predetermined in the health monitoring management plan, and use the elderly health information collection system to perform body identification inquiries and physiological health signal collection to obtain the corresponding physical health feelings and physiological health signal spectrograms of the elderly in the pension institution; A health activity anomaly analysis module, configured to perform health signal anomaly localization on the physiological health signal spectrogram corresponding to the elderly in the pension institution based on the corresponding physical health feelings of the elderly in the pension institution to obtain the corresponding health signal anomaly sites of the elderly in the pension institution; and real-time monitor the corresponding activity tracking posture status of the elderly in the pension institution through the health monitoring robot, and perform activity posture anomaly analysis according to the corresponding activity tracking posture status of the elderly in the pension institution to obtain the corresponding activity tracking posture anomaly points of the elderly in the pension institution; A remote monitoring and alarm module, configured to perform an early warning response based on the corresponding health signal anomaly sites or activity tracking posture anomaly points of the elderly in the pension institution and using the cloud control platform to generate a corresponding health anomaly early warning signal for the elderly in the pension institution, and timely send a warning message to the corresponding medical staff or family members of the pension institution.

[0020] In the embodiment of the present invention, please refer to Figure 1As shown in the figure, it is a module diagram of the elderly health monitoring robot system of the nursing home based on cloud intelligent control of the present invention. In this example, the elderly health monitoring robot system of the nursing home based on cloud intelligent control includes the following modules: S1: A health monitoring management generation module is used to obtain the historical physiological health data and historical daily activity data corresponding to the elderly in the nursing home, and generate a health monitoring management plan corresponding to the elderly in the nursing home based on the historical physiological health data and historical daily activity data and using the cloud control platform; In the embodiment of the present invention, in the nursing home, various professional equipment are used to collect the historical physiological health data and historical daily activity data of the elderly, and the electronic blood pressure meter is used to measure the blood pressure of the elderly three times a day in the morning, noon and evening. After the measurement, the device automatically stores the systolic and diastolic pressure data in the form of digital signals in the local database. The intelligent thermometer uses infrared sensing technology to automatically measure and upload the body temperature data of the elderly to the database every hour. The smart bracelet has built-in photoelectric sensors and acceleration sensors to collect heart rate and respiratory rate data at a frequency of 5 times per second, and transmits them to the local storage device via Bluetooth. At the same time, cameras are installed in various areas of the nursing home to capture the activity pictures of the elderly at a frame rate of 30 frames per second. The number of behaviors such as walking, standing, and sitting of the elderly is counted using image recognition technology and specific algorithms to calculate the amount of daily activities. The positioning bracelet containing a GPS module and an inertial measurement unit worn by the elderly records the position information and posture change data 10 times per second, and the motion trajectory is drawn accordingly. The collected data is transmitted to the cloud control platform at a speed of 1MB per second through a wired or wireless network. The platform uses a data analysis algorithm to combine the historical physiological health data and daily activity data of the elderly to generate a health monitoring management plan. For example, based on the elderly’s blood pressure, heart rate data and daily activity level, we can determine the appropriate dosage of health medication and diet plan, and plan the patrol route of the health monitoring robot based on their daily activity trajectory.

[0021] S2: A health monitoring execution and collection module, which is used to send corresponding health monitoring task instructions to the corresponding health monitoring robot based on the health monitoring management plan corresponding to the elderly in the nursing home and using the cloud control platform, and control the corresponding navigation and autonomous mobile system of the health monitoring robot to perform navigation to the location of the elderly in the nursing home based on the health monitoring task instruction response, and at the same time act on the corresponding environmental interaction system to perform medicine delivery and meal delivery tasks according to the health monitoring tasks predetermined in the health monitoring management plan, and use the elderly health information collection system to perform body recognition inquiry and physiological health signal collection, so as to obtain the corresponding physical health feelings and physiological health signal spectra of the elderly in the nursing home; In the embodiment of the present invention, the cloud control platform sends health monitoring task instructions to the corresponding health monitoring robot at a frequency of 10 times per second through the wireless communication network according to the generated health monitoring and management plan for the elderly in the elderly care institution. After receiving the instructions, the local control unit of the health monitoring robot immediately responds. On the one hand, it activates the navigation and autonomous movement system. The lidar carried by the robot scans the surrounding environment 15 times per second to obtain distance data, and the camera captures environmental images at 30 frames per second. The visual SLAM technology is used to combine the lidar data to construct a real-time environmental map, and then according to the pre-entered map of the elderly care institution and the signal of the elderly positioning bracelet, the path is planned through the triangulation algorithm, and it autonomously navigates to the location of the elderly. On the other hand, it triggers the environmental interaction system. According to the pre-determined medicine delivery and meal delivery tasks in the health monitoring and management plan, using the robotic arm with degrees of freedom and the anti-slip gripper, when arriving at the door of the elderly's room, the visual sensor is used to identify the position of the door lock, and the robotic arm is operated to open the door, accurately delivering health medicines and dining items. At the same time, the elderly health information collection system is started. The voice module sends a body recognition query to the elderly, and the speech recognition module analyzes the elderly's answer. The non-contact physiological health signal collection device, such as a millimeter wave radar sensor, collects the heart rate and respiratory rate 5 times per second, and the infrared body temperature sensor measures the body temperature. The physiological health signals are processed by the Fourier transform algorithm, and finally the corresponding physical health feelings and physiological health signal spectrograms of the elderly in the elderly care institution are obtained.

[0022] S3: The health activity abnormal analysis module is used to perform health signal abnormal positioning on the physiological health signal spectrogram corresponding to the elderly in the elderly care institution based on the corresponding physical health feelings of the elderly in the elderly care institution to obtain the health signal abnormal sites corresponding to the elderly in the elderly care institution; the activity tracking posture status corresponding to the elderly in the elderly care institution is monitored in real time through the health monitoring robot, and activity posture abnormal analysis is performed according to the activity tracking posture status corresponding to the elderly in the elderly care institution to obtain the activity tracking posture abnormal points corresponding to the elderly in the elderly care institution; In the embodiments of the present invention, for the physiological health signal spectrogram corresponding to the elderly in the elderly care institution, with the help of specialized signal analysis software, health signal anomaly localization is carried out based on the physical health feelings of the elderly. For example, if the elderly feedback dizziness, the analysis software takes the feedback time point as the center, intercepts 10 minutes of data before and after in the physiological health signal spectrogram, and compares it with the normal physiological signal fluctuation range. If it is found that the heart rate rises by more than 20 beats per minute within 1 minute, or the blood pressure drops by more than 20 mmHg within 2 minutes, etc., it is determined as a health signal anomaly site. The health monitoring robot monitors the activity tracking posture of the elderly in real time through the equipped high-definition camera and infrared sensor. For the daily activity volume, the number of actions per unit time is counted through image recognition; for the sleep body movement posture, a low-light camera is installed in the sleep area, and image analysis algorithms are used to identify actions such as turning over and kicking. According to these activity tracking posture conditions, a judgment model is set. For example, if the daily activity volume suddenly drops to zero and does not recover for 3 minutes, combined with abnormal body posture, it is judged as the sudden time point of activity fall; if the sleep body movement frequency exceeds the normal range (such as 5-10 times per hour normally, exceeding 15 times) and the actions are disorderly and violent, it is determined as an abnormal node of sleep activity, and the activity tracking posture abnormal points are summarized.

[0023] S4: The remote monitoring and alarm module is used to carry out early warning response according to the health signal anomaly sites or activity tracking posture abnormal points corresponding to the elderly in the elderly care institution and use the cloud control platform to generate a health anomaly early warning signal corresponding to the elderly in the elderly care institution, so as to send early warning information to the corresponding medical staff or family members of the elderly care institution in a timely manner.

[0024] In the embodiments of the present invention, once the health signal anomaly sites or activity tracking posture abnormal points corresponding to the elderly in the elderly care institution are detected, the cloud control platform immediately starts the early warning response mechanism. The platform has pre-set early warning rules for different abnormal situations. When receiving the abnormal information, it sends an early warning message containing the identity of the elderly and the details of the abnormal situation (such as the excessive heart rate value, the suspected fall time and place, etc.) to the mobile phone of the corresponding medical staff in the elderly care institution through the SMS interface; at the same time, it pushes an early warning notice to the mobile phone APP of the elderly's family members through the instant messaging interface, elaborating on the health abnormal condition of the elderly, generating a health anomaly early warning signal corresponding to the elderly in the elderly care institution, and ensuring that relevant personnel can be informed in a timely manner and take corresponding measures. Furthermore, the health monitoring and management generation module includes the following functions: Obtain the historical physiological health data corresponding to the elderly in the elderly care institution, including heart rate, respiratory rate, blood pressure, and body temperature; Obtain the historical daily activity data corresponding to the elderly in the elderly care institution, including daily activity volume, movement trajectory, and posture changes; Transmit the historical physiological health data and historical daily activity data corresponding to the elderly in the elderly care institution to the corresponding cloud control platform, and use the cloud control platform to perform preliminary preprocessing on the historical physiological health data and historical daily activity data, including denoising, filtering, and standardization, to remove the corresponding noise and interference information in the data. At the same time, perform personal classification annotation storage on the preprocessed historical physiological health data and historical daily activity data to generate a historical health monitoring data file corresponding to the elderly in the elderly care institution; Use the cloud control platform and combine the historical health monitoring data file corresponding to the elderly in the elderly care institution to generate a health monitoring management plan corresponding to the elderly in the elderly care institution.

[0025] As an embodiment of the present invention, refer to Figure 2 shown, for Figure 1 the functional flow diagram of the health monitoring management generation module in S11: Obtain the historical physiological health data corresponding to the elderly in the elderly care institution, including heart rate, respiratory rate, blood pressure, and body temperature; In the embodiment of the present invention, in the elderly care institution, collect the historical physiological health data corresponding to the elderly through a variety of professional medical devices. Use an electronic sphygmomanometer to regularly measure the blood pressure of the elderly. After each measurement, the device automatically stores the systolic and diastolic blood pressure data in the local database in the form of digital signals. Use an intelligent thermometer to measure the body temperature of the elderly through infrared induction technology, and the measurement results are automatically uploaded to the database every hour. For heart rate and respiratory rate, use wearable devices, such as smart bracelets, which are equipped with photoelectric sensors and acceleration sensors to collect heart rate and respiratory rate data at a frequency of 5 times per second, and transmit the data to the local storage device through Bluetooth. For example, within a week, continuously collect the physiological health data of an elderly person, and summarize these data to form a set of historical physiological health data including heart rate, respiratory rate, blood pressure, and body temperature.

[0026] S12: Obtain the historical daily activity data corresponding to the elderly in the elderly care institution, including daily activity volume, movement trajectory, and posture change; In the embodiments of the present invention, historical daily activity data corresponding to the elderly in the elderly care institution is obtained by means of a variety of monitoring devices. Cameras are installed in various areas of the elderly care institution, and image recognition technology is used to capture the activity pictures of the elderly at a frame rate of 30 frames per second. The actions of the elderly are recognized through a specific algorithm, and the number of times of behaviors such as walking, standing, and sitting of the elderly is counted, so as to calculate the daily activity amount. At the same time, the bracelet with positioning function worn by the elderly is built-in with a GPS module and an inertial measurement unit, and records the position information and posture change data of the elderly at a frequency of 10 times per second. According to this position information, the movement trajectory of the elderly can be drawn. For example, through one week of monitoring, the walking route, activity area and posture change of an elderly person in the elderly care institution every day are recorded, and the historical daily activity data is summarized, including the daily activity amount, movement trajectory and posture change information.

[0027] S13: Transmit the historical physiological health data and historical daily activity data corresponding to the elderly in the elderly care institution into the corresponding cloud control platform, and use the cloud control platform to perform preliminary preprocessing on the historical physiological health data and historical daily activity data, including denoising, filtering and standardization, so as to remove the corresponding noise and interference information in the data. At the same time, the preprocessed historical physiological health data and historical daily activity data are classified and labeled for storage to generate a historical health monitoring data file corresponding to the elderly in the elderly care institution; In the embodiments of the present invention, the collected historical physiological health data and historical daily activity data corresponding to the elderly in the elderly care institution are transmitted to the corresponding cloud control platform through a wired network or a wireless network at a transmission speed of 1MB per second. In the cloud control platform, special data processing software is used to perform preliminary preprocessing on the data. For historical physiological health data and historical daily activity data, the median filtering algorithm is used to remove the noise points in the data. For example, in blood pressure data, abnormal high or low data points caused by measurement errors are removed. The normalization method is used to standardize the data, and the physiological health data and daily activity data in different ranges are unified into the interval of 0-1 for subsequent analysis. After the preprocessing is completed, according to the personal information of the elderly, such as name, age, room number, etc., the data is classified and labeled and stored in the cloud database, and finally a historical health monitoring data file corresponding to each elderly person in the elderly care institution is generated.

[0028] S14: Use the cloud control platform and combine the historical health monitoring data file corresponding to the elderly in the elderly care institution to generate a health monitoring management plan corresponding to the elderly in the elderly care institution.

[0029] In the embodiment of the present invention, based on the historical health monitoring data file corresponding to the elderly in the pension institution, the cloud control platform generates a health monitoring management plan corresponding to the elderly in the pension institution by using data analysis algorithms. For the health monitoring autonomous navigation task, according to the daily activity trajectory and frequently occurring areas of the elderly, the best patrol path of the health monitoring robot is planned to ensure that the needs of the elderly can be discovered in time. In terms of the healthy dosage of medicine, combined with the historical physiological health data of the elderly, such as blood pressure and heart rate data, and the diagnosis suggestions of doctors, the types and dosages of medicine taken every day are determined. For the healthy diet condition, according to the age, physical condition and daily activity amount of the elderly, a personalized diet plan is formulated, including the types of food, intake, etc. In the activity tracking and monitoring management task, according to the historical posture change data and activity amount of the elderly, reasonable activity amount targets and abnormal posture warning thresholds are set. These tasks are integrated to form a complete health monitoring management plan, and the plan is sent to the corresponding health monitoring robot in the form of instructions to guide it to perform the monitoring task.

[0030] Furthermore, the health monitoring management plan corresponding to the elderly in the pension institution specifically includes corresponding health monitoring autonomous navigation, healthy dosage of medicine, healthy diet condition, and activity tracking and monitoring management tasks.

[0031] Furthermore, the health monitoring execution and collection module includes the following functions: Based on the health monitoring management plan corresponding to the elderly in the pension institution and using the cloud control platform, send the health monitoring task instructions corresponding to the health monitoring management plan to the corresponding health monitoring robot, and according to the local control unit corresponding to the health monitoring robot, receive and respond to the corresponding health monitoring task instructions to execute the health monitoring task corresponding to the elderly in the pension institution; Based on the health monitoring task instructions acting on the corresponding local control unit, respond to control the navigation and autonomous movement system corresponding to the health monitoring robot to execute autonomous navigation to the location of the elderly in the pension institution; Based on the health monitoring task instructions acting on the environment interaction system corresponding to the health monitoring robot, execute the medicine delivery and meal delivery tasks corresponding to the elderly in the pension institution according to the healthy dosage of medicine and the healthy diet condition management tasks predetermined in the health monitoring management plan, navigate autonomously to the location of the elderly in the pension institution according to the medicine delivery and meal delivery tasks, and realize the door opening and closing operation of the corresponding room at the location through the degree-of-freedom robotic arm and anti-slip gripper, and at the same time, use the degree-of-freedom robotic arm and anti-slip gripper to safely and accurately complete the transfer of the healthy medicine and dining items corresponding to the elderly in the pension institution; Use the elderly health information collection system corresponding to the health monitoring robot to conduct body recognition inquiries and physiological health signal collection to obtain the physical health feelings and physiological health signal spectrograms corresponding to the elderly in the pension institution.

[0032] As an embodiment of the present invention, refer toFigure 3 As shown in Figure 1 the functional flow diagram of the health monitoring execution and collection module in S21: Based on the health monitoring management plan corresponding to the elderly in the pension institution and using the cloud control platform, send the health monitoring task instructions corresponding to the health monitoring management plan to the corresponding health monitoring robot, and according to the local control unit corresponding to the health monitoring robot, receive and respond to the corresponding health monitoring task instructions to execute the health monitoring tasks corresponding to the elderly in the pension institution; In the embodiment of the present invention, the pension institution formulates a health monitoring management plan for each elderly person in advance. The plan clearly stipulates the specific content, time arrangement and other information of the monitoring tasks. After the cloud control platform reads the plan, it sends the corresponding health monitoring task instructions to the corresponding health monitoring robot through the wireless communication network at a frequency of 10 times per second. The local control unit of the health monitoring robot is equipped with a high-performance signal receiver to receive these instructions in real time. Once the instructions are received, the local control unit immediately parses the instructions to identify key information such as the task type and target object. For example, when receiving a task instruction to perform daily health monitoring on a certain elderly person, the local control unit quickly responds, mobilizes the internal resources of the robot, and prepares to execute the health monitoring tasks corresponding to the elderly in the pension institution.

[0033] S22: Based on the health monitoring task instructions acting on the corresponding local control unit, respond to control the navigation and autonomous movement system of the corresponding health monitoring robot to perform autonomous navigation to the location of the elderly in the pension institution; In the embodiment of the present invention, based on the received health monitoring task instructions, the local control unit activates the navigation and autonomous movement system of the health monitoring robot. The lidar carried by the robot scans the surrounding environment at a frequency of 15 times per second to obtain distance data. At the same time, the built-in camera collects environmental images at a speed of 30 frames per second. Using visual SLAM technology and combining lidar data, the surrounding environment map is constructed and updated in real time. Using the pre-entered map data of the pension institution and the signal of the positioning bracelet worn by the elderly, through the triangulation algorithm, the location of the elderly is accurately determined. The navigation system plans the optimal path from the current position of the robot to the location of the elderly according to this information. The robot autonomously navigates along the planned path by controlling the motor speed and steering, and continuously adjusts the motion state according to the sensor data during the movement to ensure accurate arrival at the location of the elderly. The algorithm plans the optimal path from the current position of the robot to the location of the elderly. The robot autonomously navigates along the planned path by controlling the motor speed and steering, and continuously adjusts the motion state according to the sensor data during the movement to ensure accurate arrival at the location of the elderly.

[0034] S23: Based on the health monitoring task instruction, the environmental interaction system corresponding to the health monitoring robot executes the medicine delivery and meal delivery tasks for the elderly in the elderly care institution according to the predetermined healthy medication dosage and the health diet management task in the health monitoring management plan, autonomously navigates to the location of the elderly in the elderly care institution according to the medicine delivery and meal delivery tasks, and realizes the door opening and closing operation of the corresponding door at the location through the robotic arm with degrees of freedom and the anti-slip gripper. At the same time, the robotic arm with degrees of freedom and the anti-slip gripper are used to safely and accurately complete the transfer of the healthy medicine and dining items for the elderly in the elderly care institution; In the embodiment of the present invention, the environmental interaction system of the health monitoring robot is triggered by the health monitoring task instruction to execute the medicine delivery and meal delivery tasks. The task instruction contains information such as the predetermined healthy medication dosage and the healthy diet status in the health monitoring management plan. The robot replans the path according to the medicine delivery and meal delivery tasks and navigates to the location of the elderly in the elderly care institution. When reaching the door of the elderly's room, the robot uses the visual sensor installed at the front end of the robotic arm to identify the position of the door lock, and uses the robotic arm with degrees of freedom and the anti-slip gripper to perform the door opening and closing operation. For example, the robotic arm extends to the door lock, the anti-slip gripper accurately clamps the doorknob, and the robotic arm is driven by the motor to rotate a certain angle to open the door. After entering the room, the robot uses the robotic arm and the anti-slip gripper to safely and accurately deliver the healthy medicine and dining items to the elderly according to the placement requirements of the medicine and dining items in the task instruction. For example, the medicine box containing the medicine is placed steadily on the table in front of the elderly, and the dinner plate is gently placed in a suitable position.

[0035] S24: Use the elderly health information collection system corresponding to the health monitoring robot to conduct body recognition inquiries and collect physiological health signals to obtain the physical health feelings and physiological health signal spectrograms of the elderly in the elderly care institution.

[0036] In the embodiment of the present invention, after the health monitoring robot completes the medicine delivery and meal delivery tasks, its elderly health information collection system is activated. The voice module in the system uses the speaker to send body recognition inquiries to the elderly in a gentle tone, such as "Excuse me, how do you feel today? Is there anywhere uncomfortable?" At the same time, the speech recognition module analyzes the elderly's answers in real time. The non-contact physiological health signal collection devices carried by the robot, such as millimeter-wave radar sensors and infrared body temperature sensors, start to work. The millimeter-wave radar sensor collects the heart rate and respiratory rate information of the elderly at a frequency of 5 times per second by transmitting and receiving millimeter waves; the infrared body temperature sensor accurately measures the body temperature by detecting the infrared radiation emitted by the elderly's body, arranges the collected physiological health signals in time series, and performs spectral conversion using the Fourier transform algorithm. Finally, the physiological health signal spectrogram of the elderly in the elderly care institution is obtained, providing data support for subsequent health condition analysis.

[0037] Further, the autonomous navigation of the health monitoring robot to the location of the elderly in the elderly care institution by the local control unit in response to the health monitoring task instruction includes: The local control unit acts on the health monitoring task instruction to respond and control the navigation and autonomous movement system of the health monitoring robot to construct a three-dimensional map of the monitoring environment of the elderly care institution, so as to accurately locate the specific location of the elderly to be monitored in the elderly care institution; In the embodiment of the present invention, the health monitoring task instruction is sent to the local control unit, and the unit immediately activates the navigation and autonomous movement system of the health monitoring robot. The lidar carried by the robot scans the environment in the elderly care institution omni-directionally at a frequency of 20 times per second to obtain a large amount of distance data point cloud information. At the same time, the built-in camera of the robot real-time collects the surrounding environment images, and uses the visual SLAM (Simultaneous Localization and Mapping) technology to combine with the lidar data to construct a three-dimensional map of the monitoring environment of the elderly care institution. During the map construction process, by comparing and correcting with the pre-entered building drawing data of the elderly care institution, the accuracy of the map is ensured. After the map construction is completed, using the positioning base stations installed in each area of the elderly care institution and the positioning-enabled bracelets worn by the elderly, through the principle of triangulation, the specific location of the elderly to be monitored in the elderly care institution is accurately located. For example, it is determined that the elderly is in the 3rd room on the east side of the second floor of the elderly care institution.

[0038] Preferably, based on the three-dimensional map of the monitoring environment and combined with the algorithm predicts the distribution of static obstacles between the location of the elderly in the elderly care institution and the health monitoring robot, and based on the distribution of static obstacles, path planning is performed on the health monitoring robot to generate the running distribution trajectory between the location of the elderly and the health monitoring robot; In the embodiment of the present invention, based on the constructed three-dimensional map of the monitoring environment, the map data is input into the path planning algorithm module, and this module uses the algorithm to perform path planning. The algorithm takes the current position of the robot as the starting point and the location of the elderly as the end point, and at the same time considers the static obstacle information in the map, such as walls, fixed furniture, etc. The algorithm calculates the cost function from each node to the starting point and the end point, and continuously searches for the path with the minimum cost. During the search process, according to the distribution of static obstacles in the map, the impassable areas are excluded. For example, when encountering a wall, the algorithm will automatically avoid this area and re-search for a feasible path. After a series of searches and calculations, the running distribution trajectory between the location of the elderly and the health monitoring robot is generated. This trajectory bypasses all static obstacles and is a theoretically feasible path.

[0039] Preferably, the current corresponding motion position, motion speed, motion direction and the distribution of real-time environmental obstacles around the health monitoring robot are obtained according to the operation distribution trajectory between the location of the elderly and the health monitoring robot, and the feasible motion window of the health monitoring robot at each moment is analyzed based on the current corresponding motion position, motion speed, motion direction and the distribution of real-time environmental obstacles around the health monitoring robot on the trajectory and in combination with the dynamic window method; In the embodiment of the present invention, the health monitoring robot moves along the previously generated operation distribution trajectory, and the sensors inside it collect the data of the current corresponding motion position, motion speed, motion direction and the distribution of real-time environmental obstacles around it on the trajectory in real time. The odometer of the robot accurately records the motion position and speed information, the gyroscope determines the motion direction, and the lidar continuously scans to obtain the distribution of real-time environmental obstacles around. These data are input into the dynamic window method analysis module, which generates multiple possible motion windows at each moment according to the kinematic model of the robot and the current motion state. The motion window refers to the area that the robot can reach within a certain time range. For example, at a certain moment, according to the current speed and acceleration limits of the robot, three motion windows are generated, corresponding to different forward directions and speed changes respectively. Each window takes into account the distribution of real-time environmental obstacles around to ensure that the robot will not collide with obstacles within the window.

[0040] Preferably, the target trajectory of the feasible motion window of the health monitoring robot at each moment is optimized based on the motion direction and the distribution of real-time environmental obstacles around, so as to determine the turning corner deviation between the health monitoring robot and the distribution of each real-time environmental obstacle around in the motion window according to the motion direction, and plan and optimize the obstacle avoidance operation trajectory corresponding to the health monitoring robot in the feasible motion window at each moment based on the corresponding turning corner deviation, so as to generate the target optimized motion trajectory corresponding to the health monitoring robot in each motion window; In an embodiment of the present invention, for the feasible motion window of the health monitoring robot at each moment, the target trajectory is optimized according to the current motion direction of the robot and the distribution of real-time environmental obstacles around it. Based on the motion direction, a geometric algorithm, such as the cosine theorem, is used to calculate the turning corner deviation between the health monitoring robot and the distribution of each real-time environmental obstacle around it within the motion window. Assume that the current motion direction of the robot is due east. If an obstacle is detected in the left front within the motion window, it is calculated that the robot needs to turn left by a certain angle to avoid the obstacle. The difference between this angle and the ideal motion direction is the turning corner deviation. Based on this deviation, the running trajectory for the health monitoring robot to avoid obstacles within the feasible motion window at each moment is planned and optimized. For example, according to the turning corner deviation, the speed and steering angle of the robot within the motion window are adjusted to generate a new running trajectory for avoiding obstacles. This trajectory enables the robot to avoid obstacles more safely and efficiently, and finally generates the target optimized motion trajectory corresponding to the health monitoring robot within each running window.

[0041] Preferably, based on the target optimized motion trajectory corresponding to the health monitoring robot within each running window, the corresponding health monitoring robot is autonomously navigated to the location of the elderly in the elderly care institution.

[0042] In an embodiment of the present invention, the health monitoring robot continuously adjusts its own motion parameters, such as speed, steering angle, etc., according to the target optimized motion trajectory corresponding to each running window to achieve autonomous navigation. The robot controls the rotation speed and steering of the motor and moves along the target optimized motion trajectory towards the location of the elderly in the elderly care institution. During the movement, it continuously monitors its own motion state and changes in the surrounding environment. If a new obstacle is found or the motion state changes, it returns to the previous step, generates a motion window again and performs optimization to ensure that the robot can accurately and safely reach the location of the elderly. For example, when approaching the room where the elderly is located, the robot detects a temporarily placed stool at the door and successfully avoids the stool by re-planning the motion trajectory and finally accurately reaches the location of the elderly to complete the autonomous navigation task.

[0043] Further, the determination of the turning corner deviation between the health monitoring robot and the distribution of each real-time environmental obstacle around it within the motion window according to the motion direction includes: A local coordinate system is established with the current position of the health monitoring robot as the origin within the motion window, and based on the motion direction, the direction position corresponding to the health monitoring robot and the distribution positions of each real-time environmental obstacle around it within the motion window are determined according to the local coordinate system; In the embodiment of the present invention, during the movement of the health monitoring robot, a square movement window with a side length of 2 meters is established with its current position as the origin. By means of the gyroscope and accelerometer inside the robot, the movement direction of the robot is determined, and a local coordinate system is constructed based on this. For example, if the robot moves due north, the due north direction is set as the positive direction of the Y-axis of the coordinate system, and the due east direction is the positive direction of the X-axis. The lidar carried by the robot scans the surrounding environment at a frequency of 10 times per second to obtain distance information. Through the principle of triangulation, each reflection point scanned by the lidar is positioned in the local coordinate system, so as to determine the distribution positions of the surrounding real-time environmental obstacles. At the same time, the robot accurately records its own direction position in the local coordinate system through its own odometer data.

[0044] Preferably, relative position angle analysis is performed according to the direction position corresponding to the health monitoring robot and the distribution positions of each surrounding real-time environmental obstacle, so as to calculate the relative position angles between the health monitoring robot and each surrounding real-time environmental obstacle by using geometric algorithms. In the embodiment of the present invention, based on the previously determined direction position corresponding to the health monitoring robot and the distribution positions of each surrounding real-time environmental obstacle, the geometric algorithm of the cosine theorem is used to calculate the relative position angle. Let the position of the robot be point A and the position of a certain obstacle be point B. According to the coordinates of two points (xA, yA) and (xB, yB) in the coordinate system, first calculate the length dAB of line segment AB, and then calculate the vectors OA (the vector from the robot position to the origin) and OB (the vector from the obstacle position to the origin) according to the coordinate axis directions determined by the movement direction of the robot. Using the cosine theorem formula cosθ = (|OA|² + |OB|² - |AB|²) / (2×|OA|×|OB|), calculate the relative position angle θ between the robot and the obstacle. For example, the robot position coordinates are (1, 0), and the coordinates of a certain obstacle are (2, 2). After calculation, the relative position angle is about 45 degrees. Repeat this calculation process for all surrounding real-time environmental obstacles, and finally obtain the relative position angles between the robot and each obstacle.

[0045] Preferably, obtain the maximum acceleration and maximum turning rate corresponding to the health monitoring robot within this movement window, and perform turning corner deviation calculation on the relative position angles between the health monitoring robot and each surrounding real-time environmental obstacle based on the maximum acceleration and maximum turning rate, so as to obtain the turning corner deviation between the health monitoring robot and the distribution of each surrounding real-time environmental obstacle within this movement window.

[0046] In the embodiments of the present invention, the maximum acceleration and maximum turning rate data in different operating states are pre-stored inside the health monitoring robot. For example, in the normal walking mode, the maximum acceleration is set to 0.5 m / s² and the maximum turning rate is 15 degrees per second. The actual acceleration and turning rate data of the robot within the current motion window are obtained and compared with the maximum acceleration and maximum turning rate. For each relative position angle, according to the kinematic model of the robot, considering the influence of acceleration and turning rate on the motion trajectory, the turning corner deviation is calculated. Assuming the robot moves at the maximum acceleration, during the turning process, according to the current relative position angle and the maximum turning rate, the ideal turning trajectory is predicted. The actual turning trajectory is compared with the ideal turning trajectory, and by calculating the angle difference between the two, the turning corner deviation between the health monitoring robot and the distribution of various surrounding real-time environmental obstacles within the motion window is obtained. For example, within a certain motion window, the relative position angle between the robot and a certain obstacle is 30 degrees. At the maximum acceleration and maximum turning rate, the angle difference between the ideal turning trajectory and the actual turning trajectory is 5 degrees, that is, the turning corner deviation corresponding to this obstacle is 5 degrees. Finally, the turning corner deviation between the health monitoring robot and the distribution of various surrounding real-time environmental obstacles within the motion window is obtained.

[0047] Further, the body recognition inquiry and physiological health signal collection using the elderly health information collection system corresponding to the health monitoring robot include: Using the elderly health information collection system corresponding to the health monitoring robot to conduct a body recognition inquiry, so as to actively inquire about the physical feelings of the elderly in the nursing home during the medicine delivery and meal delivery processes corresponding to the health monitoring robot through voice recognition, and monitor and identify the call for help and groan feeling information of the elderly in the nursing home during this period, so as to obtain the physical health feelings of the elderly in the nursing home; In the embodiments of the present invention, when the health monitoring robot executes the medicine delivery and meal delivery tasks, its built-in elderly health information collection system activates the body recognition inquiry function. The robot uses voice synthesis technology to actively inquire about the physical feelings of the elderly in a gentle and clear voice, such as "Hello, how do you feel today?" At the same time, the voice recognition module in the system starts to work. This module uses deep learning algorithms to analyze the answers of the elderly in real time. During this process, the system can also use voice monitoring technology to monitor and identify the call for help and groan feeling information emitted by the elderly. Once keywords such as "pain" and "uncomfortable" are captured, or abnormal groans are detected, the system immediately records the relevant information and integrates these information into the physical health feelings of the elderly in the nursing home. For example, the elderly answers "I'm a little dizzy today", or during this process, the system monitors that the elderly does not respond and makes abnormal groans, the system records this information and includes it in the physical health feeling data.

[0048] Preferably, the physiological health condition monitoring is carried out by using the physiological health sign monitoring in the elderly health information collection system corresponding to the health monitoring robot, so as to collect the physiological health signals of the heart rate, respiratory rate and body temperature of the elderly in the pension institution through the corresponding non-contact sensors in the health monitoring robot, and obtain the physiological health signal condition of the elderly in the pension institution; In the embodiment of the present invention, the health monitoring robot is equipped with advanced non-contact sensors for collecting the physiological health signs of the elderly. During the daily monitoring process, when the robot approaches the elderly, the infrared body temperature sensor inside it starts to work. By detecting the infrared radiation emitted by the elderly's body, the body temperature of the elderly is accurately measured, and the measurement accuracy can reach ±0.1 °C. At the same time, the millimeter-wave radar sensor emits millimeter waves and receives the reflected waves. According to the changes in the reflected waves, the heart rate and respiratory rate of the elderly are analyzed. The sensor collects these physiological health signals at a frequency of 5 times per second and transmits the data to the elderly health information collection system. The system preliminarily arranges and calibrates the collected data, removes the outliers, and finally obtains the physiological health signal condition of the elderly in the pension institution. For example, at a certain moment, the sensor collects the heart rate of the elderly as 75 beats per minute, the respiratory rate as 18 breaths per minute, and the body temperature as 36.5 °C. The system records these data as the physiological health signal condition of the elderly at this moment.

[0049] Preferably, the signal spectrum conversion is carried out according to the physiological health signal condition of the elderly in the pension institution, so as to obtain the physiological health signal spectrogram of the elderly in the pension institution.

[0050] In the embodiment of the present invention, through the physiological health signal condition of the elderly in the pension institution obtained previously, the signal spectrum conversion is carried out by using a special signal processing algorithm. First, the physiological health signals such as heart rate, respiratory rate and body temperature collected are arranged in chronological order to form time series data. Then, the fast Fourier transform (FFT) algorithm is used to convert the signal in the time domain into the signal in the frequency domain. By analyzing the signal intensity of different frequency components, the corresponding spectrogram is generated. For the heart rate signal, there are stable peaks in a certain frequency range under normal circumstances. If there is an abnormality, the peak position or intensity will change. These spectrograms are integrated together to form the physiological health signal spectrogram of the elderly in the pension institution. For example, through the FFT algorithm, the heart rate signal of a certain elderly within one hour is converted to obtain the frequency spectrum distribution within a specific frequency range, and this frequency spectrum distribution is a part of the physiological health signal spectrogram corresponding to the heart rate of this elderly during this time period.

[0051] Furthermore, the health activity abnormality analysis module includes the following functions: Locate abnormal health signals in the physiological health signal spectrogram corresponding to the elderly in the elderly care institution based on their physical health feelings, so as to obtain the abnormal health signal sites corresponding to the elderly in the elderly care institution; In the embodiment of the present invention, when a voice signal suspected of being a call for help or groan is determined through the call for help and groan feeling information corresponding to the physical health feelings recognized by voice previously, the voice analysis software is started. The software uses voice recognition technology to compare and analyze the collected voice according to the pre-recorded call for help and groan voice sample characteristics. Once it is confirmed that it is the call for help and groan voice of the elderly, the software begins to record the start time of the voice. When the voice ends, the software records the end time, so as to determine the corresponding call for help and groan time period. At the same time, the non-contact sensors in the health monitoring robot continuously collect the physiological health signals of the elderly, including heart rate, respiratory rate, body temperature, etc., and generate a physiological health signal spectrogram in the form of a time series. Based on the previously determined call for help and groan time period, the data analysis program of the robot retrieves the corresponding physiological health signal spectrogram during this time period. The program analyzes the signal changes in the spectrogram through a set algorithm. When the signal rises by more than 50% of the normal fluctuation range within a short time (such as within 1 minute), it is determined as a rapid rise time point; if the signal drops by more than 50% of the normal fluctuation range within the same time, it is a rapid drop time point; when the signal fluctuation shows a chaotic state and deviates from the normal fluctuation curve for more than 2 minutes, it is determined as an irregular oscillation time point. These time points are the abnormal health signal sites, and finally the abnormal health signal sites corresponding to the elderly in the elderly care institution are obtained.

[0052] Preferably, the activity tracking posture status corresponding to the elderly in the elderly care institution is monitored in real time by the health monitoring robot, including the daily activity volume and sleep body movement posture corresponding to the elderly in the elderly care institution; In the embodiment of the present invention, the health monitoring robot is deployed in various areas of the elderly care institution. Through its equipped high-definition camera and infrared sensor, it real-time tracks and monitors the activity tracking posture status of the elderly. The high-definition camera shoots the activity pictures of the elderly at a frame rate of 30 frames per second, and the infrared sensor is used to detect the thermal radiation of the elderly, so as to accurately perceive the position and actions of the elderly. For the daily activity volume, the robot uses image recognition technology to recognize actions such as the elderly walking, standing, sitting, falling, etc., and counts the number of actions occurring per unit time to calculate the daily activity volume. For the sleep body movement posture, a low-light camera installed in the elderly's sleep area continuously shoots the sleep state of the elderly at night. The robot uses image analysis algorithms to recognize body movement actions such as the elderly turning over and kicking, and records the body movement posture, and finally obtains the activity tracking posture status corresponding to the elderly in the elderly care institution.

[0053] Preferably, perform a temporal variation analysis on the activity tracking posture status corresponding to the elderly in the elderly care institution to generate an activity posture time variation sequence corresponding to the elderly in the elderly care institution; In the embodiment of the present invention, by transmitting the activity tracking posture status data corresponding to the elderly in the elderly care institution collected by the health monitoring robot to the corresponding data analysis system in the robot, for the daily activity volume, taking 1 hour as the time interval, count the activity volume values of the elderly in each time period to form a daily activity volume time variation sequence. For example, from 8 am to 9 am, the elderly walked 500 steps, got up 10 times, and fell 1 time, etc. Record these data in the 8 - 9 am time period. For the sleep body movement posture, also taking 1 hour as the interval, count the number of body movement actions of the elderly within each hour to generate a sleep body movement posture time variation sequence. For example, from 10 pm to 11 pm at night, the elderly turned over 8 times and kicked their legs 2 times, and record it in this time period. In this way, arrange the activity tracking posture data in different time periods in chronological order to finally generate a complete activity posture time variation sequence.

[0054] Preferably, perform an abnormal activity posture analysis based on the activity posture time variation sequence corresponding to the elderly in the elderly care institution. If it is for the time variation sequence corresponding to the daily activity volume, then statistically analyze the activity fall sudden time point corresponding to the elderly in the elderly care institution according to the time variation sequence corresponding to the daily activity volume; if it is for the time variation sequence corresponding to the sleep body movement posture, then determine the corresponding sleep body movement frequency according to the time variation sequence corresponding to the sleep body movement posture, and determine the sleep activity abnormal node for the time variation sequence corresponding to the sleep body movement posture based on the sleep body movement frequency to determine the corresponding sleep activity abnormal time node, and combine the activity fall sudden time point and the sleep activity abnormal time node corresponding to the elderly in the elderly care institution to obtain the activity tracking posture abnormal points corresponding to the elderly in the elderly care institution.

[0055] In the embodiment of the present invention, for the time variation sequence corresponding to the daily activity volume, set a fall judgment model. If at a certain moment, the activity volume of the elderly suddenly drops from the normal level to zero and there is no sign of resuming activity within the next 3 minutes, and at the same time, combined with the abnormal body posture of the elderly captured by the camera (such as the body tilt angle exceeding 60 degrees), determine that this moment is the activity fall sudden time point. For the time variation sequence corresponding to the sleep body movement posture, calculate the frequency of body movement actions within each hour. If the body movement frequency within a certain hour exceeds the normal range (for example, the normal sleep body movement frequency is 5 - 10 times per hour, when it exceeds 15 times), and the body movement actions are disorderly and violent, determine that there is a sleep activity abnormal node within this hour. Finally, summarize the activity fall sudden time point and the sleep activity abnormal node to obtain the activity tracking posture abnormal points corresponding to the elderly in the elderly care institution, providing key information for subsequent health monitoring.

[0056] Further, the abnormal health signal localization of the physiological health signal spectrogram corresponding to the elderly in the elderly care institution based on the corresponding physical health feelings of the elderly in the elderly care institution includes: Determine the corresponding call for help and groan time periods according to the call for help and groan feeling information in the corresponding physical health feelings of the elderly in the elderly care institution; In the embodiment of the present invention, through the corresponding sound collection device in the health care robot, such as a high-sensitivity microphone array, these microphone arrays can collect environmental sounds omnidirectionally, and exclude environmental noise interference through a noise reduction algorithm, focusing on capturing the sounds made by the elderly. When the microphone array receives a sound signal suspected of being a call for help and groan, the sound analysis software is activated. The software uses sound recognition technology to compare and analyze the collected sound according to the pre-recorded call for help and groan sound sample characteristics. Once it is confirmed that it is the call for help and groan sound of the elderly, the software begins to record the start time of the sound. When the sound ends, the software records the end time, so as to determine the corresponding call for help and groan time period. For example, in a certain room, the microphone array corresponding to the health care robot detects the groan of the elderly at 10:15 am, and the software starts timing. The sound stops at 10:20 am, and the software records this 5-minute time period as the call for help and groan time period, and finally determines the corresponding call for help and groan time period.

[0057] Preferably, based on the call for help and groan time periods corresponding to the elderly in the elderly care institution, the abnormal health signal localization of the physiological health signal spectrogram corresponding to the elderly in the elderly care institution is performed to determine the signal abnormal time points in the physiological health signal spectrogram during this time period, including rapid rise, rapid fall, and irregular oscillation time points, so as to obtain the abnormal health signal sites corresponding to the elderly in the elderly care institution.

[0058] In the embodiments of the present invention, physiological health signals of the elderly are continuously collected through corresponding non-contact sensors in the health monitoring robot, including heart rate, respiratory rate, body temperature, etc. These signals are used to generate a physiological health signal spectrogram in the form of a time series. Based on the previously determined call-for-help and groaning time periods, the corresponding data analysis program of the robot retrieves the physiological health signal spectrogram within this time period. The program analyzes the signal changes in the spectrogram through a set algorithm. When the signal rises by more than 50% of the normal fluctuation range within a short period (such as within 1 minute), it is determined as a rapid rise time point; if the signal drops by more than 50% of the normal fluctuation range within the same time, it is a rapid fall time point; when the signal fluctuation shows a chaotic state and deviates from the normal fluctuation curve for more than 2 minutes, it is determined as an irregular oscillation time point. These time points are the abnormal sites of the health signals. For example, within the call-for-help and groaning time period from 10:15 to 10:20, the heart rate signal spectrogram shows a situation where the heart rate rises 20 times per minute within 1 minute at 10:17, drops 15 times per minute within 1 minute at 10:18, and shows irregular oscillation from 10:19 to 10:20. Then the time points corresponding to 10:17, 10:18, and from 10:19 to 10:20 are the abnormal sites of the health signals.

[0059] Further, the remote monitoring and alarm module includes the following functions: Generate corresponding health abnormality feedback information or activity posture abnormality feedback information for the elderly in the elderly care institution according to the abnormal sites of the health signals or the abnormal points of the activity tracking postures corresponding to the elderly in the elderly care institution, and feedback and upload them to the cloud control platform; In the embodiments of the present invention, abnormal information is fed back for the abnormal sites of the health signals (i.e., the time nodes corresponding to the heart rate being higher than 120 beats per minute for 5 consecutive minutes) or the abnormal points of the activity tracking postures (i.e., the time nodes corresponding to the elderly continuously rolling over or groaning (such as for more than 3 minutes) or falling postures (the body tilt angle exceeds 60 degrees and lasts for more than 3 seconds)) determined through previous analysis. After being summarized by the signal receiver in the health monitoring robot, they are uploaded to the cloud control platform through the network, and corresponding health abnormality feedback information (such as an explanation of the heart rate abnormality) or activity posture abnormality feedback information (such as details of a suspected fall) is generated.

[0060] Preferably, the cloud control platform is used to generate a health abnormality warning signal corresponding to the elderly in the elderly care institution in response to the health abnormality feedback information or activity posture abnormality feedback information corresponding to the elderly in the elderly care institution, so as to timely send warning information corresponding to the elderly in the elderly care institution to the corresponding medical staff or family members in the elderly care institution.

[0061] In the embodiment of the present invention, after receiving the health abnormality feedback information or activity posture abnormality feedback information corresponding to the elderly in the elderly care institution through the cloud control platform, the early warning response mechanism is immediately activated. Early warning rules for different abnormal situations are preset inside the platform. For example, when receiving the health abnormality feedback information of abnormal heart rate, the platform judges the abnormality level according to the preset rules. If it is determined to be an urgent abnormality, the platform sends a warning message containing the identity of the elderly and the abnormal situation (such as the high heart rate value) to the mobile phone of the medical staff corresponding to the elderly care institution through the SMS interface; at the same time, a warning notice is pushed to the mobile phone APP of the elderly's family members through the instant messaging interface, explaining in detail the health abnormality situation of the elderly, and generating a health abnormality warning signal corresponding to the elderly in the elderly care institution, ensuring that relevant personnel can timely know the abnormal situation of the elderly and take corresponding measures.

[0062] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to cover all changes falling within the meaning and scope of the equivalent elements of the application document within the present invention.

[0063] The above are only the specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An elderly health monitoring robot system for pension institutions based on cloud intelligent control, characterized in that, It includes the following modules: A health monitoring management generation module, which is used to obtain the historical physiological health data and historical daily activity data corresponding to the elderly in the elderly care institution, and generate a health monitoring management plan corresponding to the elderly in the elderly care institution based on the historical physiological health data, historical daily activity data and using the cloud control platform; A health monitoring execution and collection module, which is used to send corresponding health monitoring task instructions to the corresponding health monitoring robot based on the health monitoring management plan corresponding to the elderly in the elderly care institution and using the cloud control platform, and based on the health monitoring task instructions, control the navigation and autonomous movement system of the corresponding health monitoring robot to navigate to the location of the elderly in the elderly care institution. At the same time, it acts on the corresponding environment interaction system to perform medicine delivery and meal delivery tasks according to the health monitoring tasks predetermined in the health monitoring management plan, and uses the elderly health information collection system to conduct body recognition inquiries and physiological health signal collection to obtain the physical health feelings and physiological health signal spectrograms corresponding to the elderly in the elderly care institution; A health activity abnormality analysis module, which is used to perform health signal abnormality localization on the physiological health signal spectrogram corresponding to the elderly in the elderly care institution based on the physical health feelings corresponding to the elderly in the elderly care institution to obtain the health signal abnormal sites corresponding to the elderly in the elderly care institution; The activity tracking posture status corresponding to the elderly in the elderly care institution is monitored in real time through the health monitoring robot, and activity posture abnormality analysis is carried out according to the activity tracking posture status corresponding to the elderly in the elderly care institution to obtain the activity tracking posture abnormal points corresponding to the elderly in the elderly care institution; A remote monitoring and alarm module, which is used to perform early warning response according to the health signal abnormal sites or activity tracking posture abnormal points corresponding to the elderly in the elderly care institution and use the cloud control platform to generate a health abnormality early warning signal corresponding to the elderly in the elderly care institution, and timely send warning information to the corresponding medical staff or family members of the elderly care institution.

2. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 1, characterized in that, The health monitoring management generation module includes the following functions: Obtain the historical physiological health data corresponding to the elderly in the elderly care institution, including heart rate, respiratory rate, blood pressure and body temperature; Obtain the historical daily activity data corresponding to the elderly in the elderly care institution, including daily activity volume, movement trajectory and posture change; Transmit the historical physiological health data and historical daily activity data corresponding to the elderly in the elderly care institution into the corresponding cloud control platform, and use the cloud control platform to perform preliminary preprocessing on the historical physiological health data and historical daily activity data, including denoising, filtering and standardization, to remove the corresponding noise and interference information in the data. At the same time, the preprocessed historical physiological health data and historical daily activity data are classified and labeled for storage to generate a historical health monitoring data file corresponding to the elderly in the elderly care institution; Use the cloud control platform and combine the historical health monitoring data file corresponding to the elderly in the elderly care institution to respond and generate a health monitoring management plan corresponding to the elderly in the elderly care institution for the corresponding health monitoring robot.

3. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 2, wherein, The health monitoring management plan corresponding to the elderly in the elderly care institution specifically includes corresponding health monitoring autonomous navigation, healthy medication dosage, healthy diet status and activity tracking monitoring management tasks.

4. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 3, characterized in that The health monitoring execution and collection module includes the following functions: Based on the health monitoring and management plan corresponding to the elderly in the elderly care institution and using the cloud control platform to send the health monitoring task instructions corresponding to the health monitoring and management plan to the corresponding health monitoring robot, and the local control unit corresponding to the health monitoring robot receives and responds to the corresponding health monitoring task instructions to execute the health monitoring tasks corresponding to the elderly in the elderly care institution; Based on the health monitoring task instructions acting on the corresponding local control unit to respond and control the navigation and autonomous movement system of the corresponding health monitoring robot to execute autonomous navigation to the location of the elderly in the elderly care institution; Based on the health monitoring task instructions acting on the environment interaction system of the health monitoring robot to execute the medicine delivery and meal delivery tasks corresponding to the elderly in the elderly care institution according to the predetermined health medication dosage and healthy diet status management tasks in the health monitoring and management plan, to autonomously navigate to the location of the elderly in the elderly care institution according to the medicine delivery and meal delivery tasks, and to realize the door opening and closing operation of the corresponding door at the location through the degree-of-freedom robotic arm and the anti-slip gripper, and at the same time use the degree-of-freedom robotic arm and the anti-slip gripper to safely and accurately complete the transfer of the health medicines and dining items corresponding to the elderly in the elderly care institution; Using the elderly health information collection system corresponding to the health monitoring robot to conduct body recognition inquiries and physiological health signal collection to obtain the physical health feelings and physiological health signal spectrograms corresponding to the elderly in the elderly care institution.

5. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 4, characterized in that, The above-mentioned based on the health monitoring task instructions acting on the corresponding local control unit to respond and control the navigation and autonomous movement system of the corresponding health monitoring robot to execute autonomous navigation to the location of the elderly in the elderly care institution includes: Based on the health monitoring task instructions acting on the corresponding local control unit to respond and control the navigation and autonomous movement system of the corresponding health monitoring robot to construct a three-dimensional map of the monitoring environment corresponding to the elderly care institution to accurately locate the specific location of the elderly to be monitored in the elderly care institution; Based on the three-dimensional map of the guardianship environment and combined with the algorithm predicts the distribution of static obstacles corresponding to the location of the elderly in the elderly care institution and the health monitoring robot, and conducts path planning for the health monitoring robot based on the distribution of static obstacles to generate the running distribution trajectory between the location of the elderly and the health monitoring robot; Obtain the current movement position, movement speed, movement direction and the distribution of surrounding real-time environmental obstacles of the health monitoring robot on the trajectory according to the running distribution trajectory between the location of the elderly and the health monitoring robot, and analyze the feasible movement window of the health monitoring robot at each moment based on the current movement position, movement speed, movement direction and the distribution of surrounding real-time environmental obstacles of the health monitoring robot on the trajectory and in combination with the dynamic window method; Based on the movement direction and the distribution of surrounding real-time environmental obstacles, optimize the target trajectory of the feasible movement window of the health monitoring robot at each moment, to determine the turning corner deviation between the health monitoring robot and the distribution of each surrounding real-time environmental obstacle in the movement window according to the movement direction, and based on the corresponding turning corner deviation, optimize the trajectory of avoiding obstacles in the feasible movement window of the health monitoring robot at each moment to generate the target optimized movement trajectory of the health monitoring robot in each movement window; Based on the target optimized movement trajectory of the health monitoring robot in each movement window, autonomously navigate the corresponding health monitoring robot to the location of the elderly in the elderly care institution.

6. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 5, characterized in that, Determining the turning corner deviation between the health monitoring robot and the distribution of each surrounding real-time environmental obstacle within the motion window according to the motion direction includes: Establishing a local coordinate system with the current position of the health monitoring robot as the origin within the motion window, and determining the corresponding direction position of the health monitoring robot and the distribution positions of each surrounding real-time environmental obstacle within the motion window based on the motion direction according to the local coordinate system; Performing relative position angle analysis based on the corresponding direction position of the health monitoring robot and the distribution positions of each surrounding real-time environmental obstacle, so as to calculate the relative position angles between the health monitoring robot and each surrounding real-time environmental obstacle by using geometric algorithms; Obtaining the maximum acceleration and maximum turning rate corresponding to the health monitoring robot within the motion window, and calculating the turning corner deviation between the health monitoring robot and each surrounding real-time environmental obstacle based on the maximum acceleration and maximum turning rate, so as to obtain the turning corner deviation between the health monitoring robot and the distribution of each surrounding real-time environmental obstacle within the motion window.

7. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 4, characterized in that, Using the elderly health information collection system corresponding to the health monitoring robot to perform body recognition inquiries and physiological health signal collection includes: Using the elderly health information collection system corresponding to the health monitoring robot to perform body recognition inquiries, so as to actively inquire about the physical feelings of the elderly in the pension institution during the medicine delivery and meal delivery processes corresponding to the health monitoring robot through voice recognition, and monitoring and recognizing the call for help and groaning feeling information of the elderly in the pension institution during this period, so as to obtain the physical health feelings of the elderly in the pension institution; Performing physiological health condition monitoring by using the physiological health sign monitoring in the elderly health information collection system corresponding to the health monitoring robot, so as to collect the heart rate, respiratory rate and body temperature physiological health signals of the elderly in the pension institution through the non-contact sensors corresponding to the health monitoring robot, and obtain the physiological health signal condition of the elderly in the pension institution; Performing signal spectrum conversion according to the physiological health signal condition of the elderly in the pension institution, so as to obtain the physiological health signal spectrogram of the elderly in the pension institution.

8. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 7, characterized in that, The health activity abnormality analysis module includes the following functions: Performing health signal abnormality localization on the physiological health signal spectrogram of the elderly in the pension institution based on the physical health feelings of the elderly in the pension institution, so as to obtain the health signal abnormal sites of the elderly in the pension institution; Real-time monitoring of the activity tracking posture status of the elderly in the pension institution by the health monitoring robot, including the daily activity volume and sleep body movement posture of the elderly in the pension institution; Performing time series change analysis on the activity tracking posture status of the elderly in the pension institution, so as to generate the activity posture time change sequence of the elderly in the pension institution. Analyze the abnormal activity postures based on the time-varying sequence of the activity postures of the elderly in the elderly care institution. For the time-varying sequence corresponding to the daily activity volume, statistically analyze the time-varying sequence corresponding to the daily activity volume to determine the sudden time point of activity falls of the elderly in the elderly care institution; for the time-varying sequence corresponding to the sleep body movement posture, determine the corresponding sleep body movement frequency based on the time-varying sequence corresponding to the sleep body movement posture, and determine the abnormal nodes of sleep activities for the time-varying sequence corresponding to the sleep body movement frequency based on the sleep body movement frequency to determine the corresponding abnormal time nodes of sleep activities, and merge the sudden time point of activity falls and the abnormal time nodes of sleep activities of the elderly in the elderly care institution to obtain the abnormal points of the activity tracking postures of the elderly in the elderly care institution.

9. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 8, characterized in that, The health signal abnormal positioning of the physiological health signal spectrogram corresponding to the elderly in the elderly care institution based on the corresponding physical health feelings of the elderly in the elderly care institution includes: Determine the corresponding call for help and groan time periods according to the call for help and groan feeling information in the corresponding physical health feelings of the elderly in the elderly care institution; Based on the call for help and groan time periods of the elderly in the elderly care institution, perform health signal abnormal positioning on the physiological health signal spectrogram corresponding to the elderly in the elderly care institution, so as to determine the signal abnormal time points in the physiological health signal spectrogram during this time period according to the call for help and groan time periods, including the rapid rise, rapid fall and irregular oscillation time points, to obtain the health signal abnormal sites corresponding to the elderly in the elderly care institution.

10. The elderly health monitoring robot system for pension institutions based on cloud intelligent control according to claim 1, characterized in that, The remote monitoring and alarm module includes the following functions: Generate the corresponding health abnormal feedback information or activity posture abnormal feedback information for the elderly in the elderly care institution according to the health signal abnormal sites or activity tracking posture abnormal points corresponding to the elderly in the elderly care institution, and feedback and upload it to the cloud control platform; Use the cloud control platform to respond to the early warning according to the health abnormal feedback information or activity posture abnormal feedback information corresponding to the elderly in the elderly care institution, generate the corresponding health abnormal early warning signal for the elderly in the elderly care institution, and timely send the warning information corresponding to the elderly in the elderly care institution to the corresponding medical staff or family members of the elderly care institution.

Images