Method for intelligently accompanying old people through robot
The physiological and environmental data of the elderly are obtained through robots, and deep learning models are used for analysis and personalized care, which solves the problems of health monitoring and environmental safety of the elderly, achieving an efficient and safe intelligent care experience.
Patent Information
- Application Number
- CN202510691450.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-27
- Publication Date
- 2025-08-05
AI Technical Summary
In the prior art, the health monitoring and daily care of the elderly have problems such as data silos, insufficient awareness of environmental security, lack of personalized interaction strategies and limited model updates, resulting in high costs, delayed response and stiff care experience.
The physiological parameters and environmental perception data of the elderly are obtained through robots, and deep learning models that combine convolutional neural networks and long-term memory networks are analyzed. Personalized accompanying instructions are generated, and voice interaction, action assistance and emergency calls are realized through online adaptive optimization models.
Accurate health risk assessment and environmental safety assessment of the elderly have been achieved, emergency response time has been shortened, naturalness and comfort of the accompanying experience, and privacy data security has been ensured.
Smart Images
Figure CN120422237A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent care technology, and in particular to a method for providing intelligent care to the elderly through robots. Background Art
[0002] The demand for health monitoring and daily care for the elderly is increasing. The traditional manual care model relies on nurses or caregivers to be on duty on site, which is not only costly but also has problems such as insufficient human resources, blind spots in monitoring, and delayed responses. Some nursing homes have introduced service robots, but they mostly focus on simple services such as meal delivery and guidance. They lack real-time monitoring and intelligent analysis of changes in the elderly's vital signs, and it is difficult to perceive environmental risks in a timely manner, and cannot provide "intelligent care" in the true sense. Existing technologies still have the following shortcomings in physiological parameter collection, environmental perception, health risk warning and autonomous decision-making:
[0003] Data silos and delayed analysis – Wearable devices often only store local physiological data and lack a unified online analysis platform, making it difficult to trigger emergency responses in a timely manner.
[0004] Insufficient awareness of environmental safety – Traditional robots rely on a single camera, making it difficult to accurately perceive the distribution of indoor obstacles and spatial layout, posing a collision risk.
[0005] Lack of personalized interaction strategies – Most systems can only execute preset instructions, ignoring the daily habits and emotional needs of the elderly, resulting in a rigid care experience;
[0006] Limited model updates - AI models trained offline are difficult to quickly optimize based on actual usage feedback, resulting in insufficient adaptability and robustness. Summary of the Invention
[0007] In order to overcome the above-mentioned problems of the related art at least to a certain extent, the present application provides a method for providing intelligent care for the elderly through robots.
[0008] The scheme of this application is as follows:
[0009] A method for providing intelligent care for the elderly using a robot, comprising:
[0010] Acquiring physiological parameter data and environmental perception data of the elderly;
[0011] Inputting the physiological parameter data and environmental perception data into a pre-trained artificial intelligence model, wherein the artificial intelligence model is used to analyze the health status of the elderly and the safety of the surrounding environment;
[0012] Generate accompanying instructions based on the analysis results of the artificial intelligence model;
[0013] The robot executes the accompanying instructions to achieve voice interaction, movement assistance or emergency calls for the elderly;
[0014] The artificial intelligence model is adaptively optimized online based on feedback data during the operation of the robot.
[0015] Preferably, the obtaining of physiological parameter data includes collecting the elderly's heart rate, blood pressure and body movement information in real time through wearable devices.
[0016] Preferably, the environmental perception data includes indoor space layout and obstacle distribution information obtained by multi-cameras and lidar.
[0017] Preferably, the artificial intelligence model is a deep learning model that integrates convolutional neural networks and long short-term memory networks, which is used to perform feature extraction and health risk prediction on image data and physiological time series data respectively.
[0018] Preferably, generating a care instruction includes:
[0019] When the health status analysis result indicates that the risk of falling is higher than a preset threshold, an emergency call instruction is generated to stop the robot movement and make a voice call for help;
[0020] When the environmental safety analysis result indicates that an obstacle blocks the passage, a motion control instruction for replanning a safe path is generated.
[0021] Preferably, when the robot executes the accompanying instructions, it also includes a personalized interaction strategy based on the daily behavioral habits of the elderly to improve the naturalness and comfort of the accompanying experience.
[0022] Preferably, the online adaptive optimization includes:
[0023] Upload sensor feedback and subjective evaluation annotations of the elderly during the robot's execution to the cloud server;
[0024] Update the AI model parameters based on the federated learning framework on the cloud server and synchronize them to the local robot.
[0025] Preferably, it also includes:
[0026] The physiological parameters and environmental perception data are encrypted and stored locally on the robot, and end-to-end encryption is used when the data is transmitted to the cloud to protect the privacy and security of the elderly.
[0027] The technical solution provided by this application may have the following beneficial effects:
[0028] This application uses multimodal sensors to capture elderly individuals' physiological parameters and environmental data in real time, achieving comprehensive awareness of heart rate, blood pressure, body movement, and spatial obstacles, eliminating the blind spots typically associated with a single data source. Using a pre-trained deep learning model to collaboratively analyze physiological and environmental data, the system accurately assesses health risks and safety hazards, generating scenario-specific care instructions, enhancing the scientificity and accuracy of decision-making. When an emergency situation, such as a fall risk or obstructed passage, is detected, the robot can immediately halt its movement and initiate voice assistance or re-plan its route, significantly reducing response time and improving emergency response efficiency. Online adaptive optimization based on feedback data during operation enables the AI model to continuously incorporate annotated information from real-world scenarios, enhancing its adaptability to individual differences and environmental changes. The generated care instructions not only include basic health monitoring and safety measures but also incorporate the elderly's daily behaviors, enabling highly personalized voice interaction and action-assisted guidance, making the care process more natural and comfortable. Local encrypted storage and end-to-end transmission effectively safeguard the privacy of the elderly's data.
[0029] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0031] Figure 1 This is a flow chart of a method for providing intelligent care for the elderly using a robot, provided in one embodiment of the present application. DETAILED DESCRIPTION
[0032] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
[0033] Figure 1 This is a flow chart of a method for providing intelligent care for the elderly by a robot, provided in one embodiment of the present application. Figure 1 , a method for providing intelligent care for the elderly by using a robot, comprising:
[0034] Obtain physiological parameter data and environmental perception data of the elderly;
[0035] Input physiological parameter data and environmental perception data into a pre-trained artificial intelligence model, which is used to analyze the health status of the elderly and the safety of the surrounding environment;
[0036] Generate accompanying instructions based on the analysis results of the artificial intelligence model;
[0037] The robot can execute accompanying instructions to achieve voice interaction, movement assistance or emergency calls for the elderly;
[0038] Perform online adaptive optimization of the artificial intelligence model based on feedback data during the robot's operation.
[0039] It should be noted that obtaining physiological parameter data includes collecting the elderly's heart rate, blood pressure and body movement information in real time through wearable devices.
[0040] In this embodiment, "obtaining physiological parameter data" specifically relies on wearable devices (such as smart bracelets, chest-mounted ECG patches, etc.) to collect key vital signs of the elderly such as heart rate, blood pressure, and body movement (exercise, posture changes) in real time.
[0041] These physiological timing data are transmitted to the robot control unit via a wireless communication module (Bluetooth / Wi-Fi) and synchronously input into the artificial intelligence model for analysis.
[0042] Compared to periodic or manual measurements, continuous monitoring can detect sudden physiological abnormalities (such as transient arrhythmias), significantly reducing the delay in detecting abnormalities. The synergy of multiple physiological parameters avoids misjudgment based on a single indicator and improves the accuracy of health risk assessment. Wearable devices are lightweight and comfortable, minimizing disruption to the daily lives of older adults and increasing system acceptance.
[0043] It should be noted that environmental perception data includes indoor space layout and obstacle distribution information obtained through multi-cameras and lidar.
[0044] The environmental perception part introduces "multi-camera + lidar" dual-modal perception:
[0045] Multi-cameras are responsible for acquiring color and depth images for 3D reconstruction of indoor spaces and detection of dynamic targets (pedestrians, pets).
[0046] LiDAR provides high-precision point clouds to compensate for the blind spots of cameras in low-light or obstructed conditions.
[0047] Vision and laser technology complement each other, enabling the robot to accurately perceive obstacles and spatial layout, even in complex environments like low light and smoke. High-precision point cloud and deep image fusion generate refined maps for real-time path planning, reducing collision risk. Supporting simultaneous perception of static and dynamic obstacles enhances the companion robot's adaptability to diverse scenarios, including those with multiple people and pets.
[0048] It should be noted that the artificial intelligence model is a deep learning model that integrates convolutional neural networks and long short-term memory networks, which is used to extract features and predict health risks for image data and physiological time series data respectively.
[0049] The artificial intelligence model uses a deep learning architecture of "Convolutional Neural Network (CNN) + Long Short-Term Memory Network (LSTM)":
[0050] The CNN branch specifically processes images or depth maps generated by cameras / radars for extracting spatial features and action recognition.
[0051] The LSTM branch performs trend analysis and anomaly detection on time series data such as heart rate, blood pressure, and body movement collected by wearable devices.
[0052] After the two features are combined in the fusion layer, health risk prediction and environmental safety assessment are output.
[0053] CNN and LSTM work together to extract features from both images and time series signals, improving recognition of complex patterns. Deep models can exploit implicit correlations in spatiotemporal data. For example, when unsteady gait (images) and heart rate fluctuations (time series) coexist, fall risk prediction becomes more reliable. This model framework supports the subsequent integration of additional sensors (such as voice, temperature, and humidity), facilitating iterative functional upgrades.
[0054] It should be noted that generating a care instruction includes:
[0055] When the health status analysis result indicates that the risk of falling is higher than the preset threshold, an emergency call instruction is generated to stop the robot movement and make a voice call for help;
[0056] When the environmental safety analysis results indicate that an obstacle blocks the passage, motion control instructions are generated to replan a safe path.
[0057] "Generate accompanying instructions" regularization:
[0058] If the model outputs "fall risk > threshold", the instruction is "stop moving + voice help (automatically call family members or nursing center)".
[0059] If the "environmental obstruction" alarm is output, the instruction is to "replan the path" and combine the SLAM algorithm to avoid obstacles and continue navigation.
[0060] It should be noted that when the robot executes care instructions, it also includes personalized interaction strategies based on the daily behavioral habits of the elderly to improve the naturalness and comfort of the care experience.
[0061] When executing accompanying instructions, the robot dynamically adjusts the voice interaction content and body assistance movements based on the elderly person's "daily behavioral habit profile" (such as preference for walking in slow mode, habitual stopping locations, common greetings, etc.).
[0062] It should be noted that online adaptive optimization includes:
[0063] Upload sensor feedback and subjective evaluation annotations of the elderly during the robot's execution to the cloud server;
[0064] Update the AI model parameters based on the federated learning framework on the cloud server and synchronize them to the local robot.
[0065] Online Adaptive Optimization consists of two layers:
[0066] Local → Cloud: The robot uploads sensor feedback (such as false alarm and missed alarm logs) and the elderly’s subjective evaluation (through interactive interface scoring or voice evaluation) to the cloud.
[0067] Cloud-based federated learning: While ensuring privacy, each robot aggregates gradients to update the model in the cloud through federated learning. After training is completed, the new parameters are sent to the local device.
[0068] The model continues to evolve without the need for manual retraining or on-site parameter adjustment. Feedback from local users can be incorporated into the next round of model updates.
[0069] The dual optimization of individual and group not only retains local fine-tuning for individual elderly people, but also absorbs the common experience of equipment across the entire network to improve the overall algorithm performance.
[0070] Data privacy protection: Federated learning avoids uploading raw data in plain text, complying with privacy compliance requirements such as GDPR.
[0071] It should be noted that the method also includes:
[0072] Physiological parameters and environmental perception data are encrypted and stored locally on the robot, and end-to-end encryption is used when the data is transmitted to the cloud to protect the privacy and security of the elderly.
[0073] It can be understood that the same or similar parts of the above embodiments can be referenced to each other, and the contents not described in detail in some embodiments can refer to the same or similar contents in other embodiments.
[0074] It should be noted that, in the description of this application, the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. In addition, in the description of this application, unless otherwise specified, the meaning of "plurality" refers to at least two.
[0075] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.
[0076] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0077] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.
[0078] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0079] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.
[0080] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
[0081] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. A method for providing intelligent care for the elderly using a robot, characterized in that: include: Acquiring physiological parameter data and environmental perception data of the elderly; Inputting the physiological parameter data and environmental perception data into a pre-trained artificial intelligence model, wherein the artificial intelligence model is used to analyze the health status of the elderly and the safety of the surrounding environment; Generate accompanying instructions based on the analysis results of the artificial intelligence model; The robot executes the accompanying instructions to achieve voice interaction, movement assistance or emergency calls for the elderly; The artificial intelligence model is adaptively optimized online based on feedback data during the operation of the robot.
2. The method according to claim 1, characterized in that The acquisition of physiological parameter data includes real-time collection of the elderly's heart rate, blood pressure and body movement information through wearable devices.
3. The method according to claim 1, characterized in that The environmental perception data includes indoor space layout and obstacle distribution information obtained through multi-cameras and lidar.
4. The method according to claim 1, wherein The artificial intelligence model is a deep learning model that integrates convolutional neural networks and long short-term memory networks, and is used to extract features and predict health risks for image data and physiological time series data respectively.
5. The method according to claim 1, characterized in that Generating the accompanying instruction includes: When the health status analysis result indicates that the risk of falling is higher than a preset threshold, an emergency call instruction is generated to stop the robot movement and make a voice call for help; When the environmental safety analysis result indicates that an obstacle blocks the passage, a motion control instruction for replanning a safe path is generated.
6. The method according to claim 1, characterized in that When the robot executes the accompanying instructions, it also includes personalized interaction strategies based on the daily behavioral habits of the elderly to improve the naturalness and comfort of the accompanying experience.
7. The method according to claim 1, characterized in that The online adaptive optimization includes: Upload sensor feedback and subjective evaluation annotations of the elderly during the robot's execution to the cloud server; Update the AI model parameters based on the federated learning framework on the cloud server and synchronize them to the local robot.
8. The method according to claim 1, characterized in that Also includes: The physiological parameters and environmental perception data are encrypted and stored locally on the robot, and end-to-end encryption is used when the data is transmitted to the cloud to protect the privacy and security of the elderly.