Ageing-suitable intelligent bed APP system based on large language model

Abstract

The invention discloses an intelligent bed APP system suitable for aging based on a large language model, and belongs to the technical field of intelligent home control and intelligent health old-age care. The system comprises a user interaction layer, a multi-mode perception layer, a heterogeneous data layer, an intelligent algorithm layer and a closed-loop feedback layer, physiological state data of a user are collected through a non-intrusive sensor, and combined analysis is performed on touch or voice input of the user and the physiological state data based on a large language model. And a structured control instruction is generated to drive a bed body execution mechanism and environment equipment to execute corresponding operation, so that self-adaptive adjustment of the bed body state and comfort parameters is realized. The intelligent bed system improves the interaction intelligence level and the use experience suitable for the aged, and is suitable for home-based care for the aged and intelligent health management scenes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart home control and wisdom health and pension, and particularly relates to an old-age-adapted intelligent bed APP system based on a large language model. BACKGROUND

[0002] With the deepening of population aging, the demand for home-based care and wisdom care continues to grow. As an important device for the daily life and health management of the elderly, the function of the intelligent bed has gradually developed from single mechanical adjustment to sleep monitoring, health assistance and environmental linkage control.

[0003] The existing intelligent bed supporting application program mostly uses fixed instructions or simple rules for human-computer interaction, which is difficult to accurately understand natural language expression and has poor voice interaction experience. At the same time, there is a lack of effective linkage between physiological state data collection, analysis and device control, and the system mostly stays at the data display level, which is difficult to realize adaptive adjustment according to the real-time state of the user. In addition, the old-age-adapted design of the existing product mostly focuses on the interface layer, and a complete intelligent decision-making and feedback loop has not been formed at the system architecture level, which affects the practicality and intelligent level of the intelligent bed system.

[0004] Therefore, it is necessary to provide an old-age-adapted intelligent bed APP system based on a large language model to improve the interaction intelligent level and old-age-adapted use experience of the intelligent bed system. SUMMARY

[0005] (I) Invention purpose The purpose of the present application is to provide an old-age-adapted intelligent bed APP system based on a large language model, which realizes accurate understanding of user needs and adaptive adjustment of the bed body and environmental devices by constructing a closed-loop technical architecture combining multi-modal perception, semantic understanding, intelligent decision-making and device control.

[0006] (II) Technical solutions To achieve the above purpose, the present application adopts the following technical solutions.

[0007] An old-age-adapted intelligent bed APP system based on a large language model adopts a layered architecture, including a user interaction layer, a multi-modal perception layer, a heterogeneous data layer, an intelligent algorithm layer and a closed-loop feedback layer.

[0008] The user interaction layer is used to receive touch input and / or voice input of the user and output operation feedback information to the user; The multi-modal perception layer is used to collect physiological state data and behavior data of the user during bed time through non-invasive sensors; The heterogeneous data layer is used to store and manage the collected multi-source heterogeneous data; The intelligent algorithm layer is used for joint analysis of user input information and physiological state data based on a large language model, and generation of structured control instructions. The closed-loop feedback layer is used for driving the bed body actuator and or the environmental equipment to perform corresponding operations according to the structured control instructions, and feeding back the execution results to the user.

[0009] (Three) beneficial effects Compared with the prior art, the present application has at least the following beneficial effects: Improve the system's understanding of natural language requirements of users; Realize the closed-loop linkage between physiological data and device control; Improve the adaptive aging interaction experience and intelligent control level of the intelligent bed system. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 is a layered architecture diagram of the adaptive aging intelligent bed APP system based on a large language model of the present application; Figure 2 is a timing diagram of the multi-modal interaction and intelligent decision-making process of the present application; Figure 3 is a data organization structure diagram of the heterogeneous data layer of the present application; Figure 4 is a main interface diagram of the adaptive aging intelligent bed APP of the present application; Figure 5 is a personalized sleep report interface diagram of the present application. DETAILED DESCRIPTION

[0011] The specific embodiments of the adaptive aging intelligent bed APP system based on a large language model of the present application will be described below with reference to the accompanying drawings.

[0012] (I) System implementation As shown in Figure 1 The adaptive aging intelligent bed APP system 100 provided by the present embodiment adopts a layered architecture, including a user interaction layer 101, a multi-modal perception layer 102, a heterogeneous data layer 103, an intelligent algorithm layer 104 and a closed-loop feedback layer 105.

[0013] The user interaction layer 101 includes a graphical interaction module 1011 and a voice interaction module 1012, which are used to receive user touch operation instructions and voice input information, and to show the system state and execution results to the user.

[0014] The multi-modal perception layer 102 includes a piezoelectric sensor 1021, an infrared sensor 1022 and a piezoelectric film sensor 1023, which are used to collect physiological state data and behavior data of the user during bed.

[0015] The heterogeneous data layer 103 includes a time-series database module 1031 and a graph database module 1032, which are used to store and manage continuous physiological state data and their relationships.

[0016] The intelligent algorithm layer 104 includes a speech recognition module 1041 and a semantic understanding and decision-making module 1042, which are used to jointly analyze user input information and physiological state data based on a large language model to generate structured control instructions.

[0017] The closed-loop feedback layer 105 includes a bed execution module 1051, an environmental control module 1052, and a feedback display module 1053, which are used to execute control operations according to the structured control instructions and provide feedback on the execution results to the user.

[0018] (II) Implementation Method like Figure 2 As shown, the interaction and control method of an age-friendly smart bed provided by the present invention includes the following steps.

[0019] In step 201, the system receives user input, which includes touch operation commands and voice input signals.

[0020] In step 202, the system collects physiological and behavioral data of the user during bedtime through the multimodal perception layer and stores the data in the heterogeneous data layer.

[0021] In step 203, the system parses and processes the user input information and physiological state data, including performing speech recognition on the voice input signal and converting it into text instructions.

[0022] In step 204, the system performs joint processing on the parsing results based on the large language model to generate structured control instructions. The structured control instructions include at least the target device identifier, the control action type, and the corresponding control parameters.

[0023] In step 205, the system drives the bed actuator and / or environmental equipment to perform corresponding control operations according to the structured control instructions.

[0024] In step 206, the system will feed back the execution results and system status information to the user, thus completing a complete interaction and control process.

[0025] Explanation of reference numerals in the attached figures 100—Age-Friendly Smart Bed APP System; 101—User Interaction Layer; 1011 – Graphical Interaction Module; 1012 — Voice interaction module; 102—Multimodal Sensing Layer; 1021—Piezoelectric sensor; 1022—Infrared sensor; 1023—Piezoelectric thin film sensor; 103—Heterogeneous Data Layer; 1031—Time Series Database Module; 1032 — Graph Database Module; 104 — Intelligent Algorithm Layer; 1041 — Speech recognition module; 1042—Semantic Understanding and Decision-Making Module; 105—Closed-loop feedback layer; 1051—Bed execution module; 1052—Environmental Control Module; 1053 — Feedback display module; 201 — User input steps; 202—Steps for collecting physiological state data; 203—Speech recognition and data parsing steps; 204 — Steps for generating structured control instructions; 205—Equipment execution steps; 206—Execution result feedback steps; 301—Physiological state data; 302 – Behavioral Data; 303 — Time Series Database Module; 304 — Graph Database Module; 305 — User profile and relationship data; 401 — Main interface display area; 402—Bed control area; 403 — Voice interaction entry point; 404 – Health information display area; 501—Sleep Cycle Structure Module; 502 – Respiration and Heart Rate Trend Display Module; 503 – Physical Activity Statistics Module; 504 - Health Tips Information Module.

Claims

1. An age-friendly smart bed APP system based on a large language model, characterized in that, include: The user interaction layer is used to receive user input and output operation feedback information to the user. The multimodal perception layer is used to collect physiological and behavioral data of users while in bed; A heterogeneous data layer is used for storing and managing the physiological state data and behavioral data; The intelligent algorithm layer is used to jointly parse the user input and the physiological state data based on a large language model to generate structured control instructions; A closed-loop feedback layer is used to drive the bed actuator and / or environmental equipment to perform corresponding operations according to the structured control instructions, and to feed back the execution results to the user interaction layer; The structured control instructions include at least the target device identifier, the control action type, and the corresponding control parameters.

2. The system according to claim 1, characterized in that, The user input includes touch operation commands and / or voice input signals.

3. The system according to claim 1, characterized in that, The multimodal sensing layer includes a pressure sensor, an infrared sensor, and / or a piezoelectric thin film sensor for collecting physiological state data.

4. The system according to claim 1, characterized in that, The heterogeneous data layer includes a time-series database module for storing continuous physiological state data, and a graph database module for storing user profiles and data relationships.

5. The system according to claim 1, characterized in that, The intelligent algorithm layer includes a speech recognition module and a semantic understanding and decision-making module, which are used to convert speech input signals into text instructions and generate the structured control instructions.

6. The system according to claim 1, characterized in that, The bed actuator includes a bed posture adjustment device, a massage device, and / or an assistive standing device.

7. The system according to claim 1, characterized in that, The feedback includes displaying the execution results in the user interface and / or outputting execution status information via voice.

8. An interaction and control method for an age-friendly smart bed, characterized in that, Includes the following steps: Receive user input; Collect physiological and behavioral data of users while in bed; Based on a large language model, the user input and the physiological state data are jointly parsed to generate structured control instructions; The structured control commands drive the bed actuators and / or environmental devices to perform corresponding control operations. The execution results will be fed back to the user.

9. The method according to claim 8, characterized in that, The user input includes touch operation commands and / or voice input signals.

10. The method according to claim 8, characterized in that, The structured control instructions include at least the target device identifier, the control action type, and the corresponding control parameters.

Images