Intelligent seat interaction method and system that fits human back

Abstract

The present invention relates to the field of intelligent seats and human-computer interaction technology, and in particular to an intelligent seat interaction method and system that fits the back of a human body. By integrating intelligent visual sensors, multi-frequency radio frequency identification, intelligent material technology, nanosensor networks, adaptive neural network models and multimodal feedback systems, an intelligent seat interaction method that is highly fitted to the back of a human body is realized; its core process includes: collecting user facial and posture features to generate personalized adaptation data, using intelligent materials and sensor networks to adjust the seat shape in real time, optimizing posture parameters through neural networks, and finally collaboratively optimizing the seat configuration through a multi-agent system; the present invention significantly improves the dynamic adaptability and personalized experience of intelligent seats, effectively solves the shortcomings of the prior art in lacking real-time feedback and personalized adaptation, and provides a higher level of comfort and health support.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent seats and human-computer interaction, and in particular to an interaction method and system for an intelligent seat that fits closely to a human back. Background Art

[0002] In daily life, the comfort of seats has always been an important demand of users. The intelligent seat interaction system that fits the back of the human body not only improves the user's comfort experience, but also promotes the user's healthy sitting posture through its dynamic adjustment function. This technology is of great significance in improving the fit between the seat and the user's back, especially in applications in offices, homes and public places, which can effectively reduce the discomfort and health problems caused by long-term sitting.

[0003] Existing smart seat solutions mainly rely on mechanical components and basic sensor technologies to achieve interaction with users.

[0004] For example, in prior art 1 (Chinese invention patent, publication number: CN111407110A, name: a seat adjustment method, device, electronic device and medium), the seat system controls the movement of the back fitting module through preset control logic. This method lacks the ability to respond to real-time changes of users and is difficult to provide a personalized comfort experience.

[0005] In prior art 2 (Chinese invention patent, publication number: CN113925311B, name: A fully automatic intelligent seat interaction method and system that fits the human back), the seat uses weight detection and thrust value matching technology to identify the user's identity and adjust the seat position, but this method mainly relies on static data and cannot dynamically adapt to the user's posture changes and needs.

[0006] Due to the lack of intelligent feedback mechanisms and adaptive capabilities, these technologies are unable to meet the personalized needs of different users, resulting in insufficient comfort and user satisfaction. Summary of the invention

[0007] In view of the many problems existing in the above-mentioned prior art, the present invention provides an interactive method and system for an intelligent seat that fits well with the human back. The present invention integrates advanced sensing technology, intelligent materials and adaptive algorithms to achieve a high degree of fit between the intelligent seat and the user's back. The user's posture and feature information are collected by intelligent visual sensors, and combined with the dynamic adjustment capability of intelligent materials and the adaptive optimization function of neural networks, the seat can respond to the user's posture changes in real time, and continuously optimize the comfort and support strength, thereby providing a personalized sitting experience.

[0008] The method for interacting with a smart seat that fits the back of a human body comprises the following steps:

[0009] The user's facial expression features, posture features and body shape features are collected through intelligent visual sensors, and the user's identity information is obtained through a multi-frequency radio frequency identification system to generate personalized adaptation data to preliminarily configure the shape of the smart seat;

[0010] According to the personalized adaptation data, the user's back pressure data is collected in real time using smart material technology and nanosensor networks, pressure feature data is generated through a multi-scale modeling algorithm, and pressure adjustment data is generated using a recursive fuzzy control algorithm to dynamically adjust the shape of the smart material;

[0011] Collect the user's sitting posture changes and micro-movements to generate behavioral feature data, and generate behavioral pattern data through time series analysis; use the adaptive neural network model to generate posture optimization parameters, and combine with the multimodal feedback system to adjust the posture of the smart seat in real time;

[0012] Integrate personalized adaptation data, pressure adjustment data and posture optimization parameters, generate collaborative optimization data through collaborative analysis of multi-agent systems, and perform self-adjustment and optimization evaluation of the intelligent seat system.

[0013] Preferably, the intelligent visual sensor includes multiple camera modules, and the camera modules are configured to capture the user's facial expression features, posture features and body features from multiple perspectives to generate personalized adaptation data, wherein the personalized adaptation data is used to adjust the initial form of the smart seat to make it more closely fit the user's body contour.

[0014] Preferably, the multi-frequency RFID system obtains the user's identity information by detecting the RFID tag worn by the user. The multi-frequency RFID system is used to monitor and update the changes of the user's identity information in real time, generate dynamic personalized adaptation data, and continuously adjust the configuration parameters of the smart seat to match the user's current identity.

[0015] Preferably, the multi-scale modeling algorithm is used to analyze the pressure data of the user's back to generate pressure characteristic data, wherein the pressure characteristic data is calculated by decomposing the real-time pressure data into multiple scales and performing weighted summation, and the expression is as follows:

[0016]

[0017] in, Indicates at time Time, location Pressure characteristic data at is the weight for each scale; is the spatial basis function; is the time basis function; is the scale quantity.

[0018] Preferably, the recursive fuzzy control algorithm generates pressure adjustment data based on the pressure characteristic data, and the pressure adjustment data is used to dynamically adjust the shape of the smart material. The recursive fuzzy control algorithm calculates an adjustment path by recursively evaluating the difference between the pressure characteristic data and the target pressure distribution.

[0019] Preferably, the behavior characteristic data includes the user's sitting posture change amplitude, micro-movement frequency and electromyographic signal change, and the behavior characteristic data is used to analyze the user's behavior characteristics to identify the user's long-term posture pattern and usage habits.

[0020] Preferably, the time series analysis analyzes the behavior feature data by short-term Fourier transform to generate behavior pattern data, and the behavior pattern data captures the dynamic characteristics and periodic characteristics of the user's posture changes.

[0021] Preferably, the adaptive neural network model generates posture optimization parameters by analyzing the behavior pattern data, and the calculation process is implemented by the following expression:

[0022]

[0023] in, Optimize parameters for posture; Represents a neural network model; Behavioral pattern data; is the time step, which is used to guide the real-time adjustment of the intelligent seat posture.

[0024] Preferably, the collaborative analysis of the multi-agent system involves information exchange and joint decision-making among multiple intelligent agents. The intelligent agents are responsible for analyzing different data dimensions in the standardized data set and generating collaborative optimization data through comprehensive analysis, thereby realizing self-adjustment and optimization evaluation of the intelligent seat system.

[0025] A system for implementing the method for interacting with a smart chair that fits the back of a human body, comprising:

[0026] An intelligent visual sensor module, configured to collect facial expression features, posture features and body shape features of a user, and generate visual feature data to provide user personalized adaptation data for preliminary configuration of the intelligent seat;

[0027] Multi-frequency radio frequency identification module, used to obtain the user's identity information and generate personalized adaptation data to ensure that the configuration of the smart seat matches the user's identity;

[0028] a smart material assembly including a flexible material configured to dynamically respond to pressure modulation data to adjust the form of the smart seat to achieve a dynamic fit to a user's back;

[0029] Nanosensor network, embedded in the smart chair, is used to collect real-time pressure data on the user's back and generate real-time pressure data for processing by multi-scale modeling algorithms;

[0030] A multi-scale modeling processing unit, used to process the real-time pressure data generated by the nanosensor network and generate pressure characteristic data through a multi-scale modeling algorithm;

[0031] a recursive fuzzy controller for receiving pressure characteristic data and generating pressure adjustment data through a recursive fuzzy control algorithm to guide dynamic adjustment of the smart material component;

[0032] The behavior analysis unit is used to collect the user's sitting posture changes and micro-movements, generate behavior feature data, and generate behavior pattern data through time series analysis;

[0033] An adaptive neural network model for processing behavioral pattern data and generating posture optimization parameters to adjust the posture of the smart seat;

[0034] Multimodal feedback system, used to provide real-time user feedback to help adjust the posture optimization of smart seats;

[0035] A multi-agent system is used to integrate personalized adaptation data, pressure regulation data, and posture optimization parameters, generate collaborative optimization data through collaborative analysis, and perform self-adjustment and optimization evaluation of the intelligent seat system.

[0036] Compared with the prior art, the advantages and beneficial effects of the present invention are:

[0037] The present invention realizes accurate identification and dynamic adaptation of user identity and characteristics through intelligent visual sensors and multi-frequency radio frequency identification system technology;

[0038] The present invention realizes real-time monitoring and feedback control of the user's back pressure through intelligent material technology and nano-sensor network technology, thereby enhancing the dynamic adaptability of the seat.

[0039] The present invention realizes real-time optimization and personalized adjustment of seat posture through adaptive neural network model and multi-modal feedback system technology;

[0040] The present invention realizes self-adjustment and optimization evaluation of the seat system through the collaborative analysis technology of the multi-agent system, which significantly improves the user's comfort and personalized experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of the process of the present invention;

[0042] Figure 2 A schematic diagram of user feature identification in the present invention;

[0043] Figure 3 A schematic diagram of dynamic adjustment of smart materials in the present invention;

[0044] Figure 4 It is a flow chart of behavior feature analysis in the present invention;

[0045] Figure 5 It is a structural block diagram of the system of the present invention. DETAILED DESCRIPTION

[0046] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is obvious that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0047] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0048] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0049] like Figure 1-Figure 2 As shown, the interaction method of the intelligent seat that fits the back of the human body includes the following steps:

[0050] The user's facial expression features, posture features and body shape features are collected through intelligent visual sensors, and the user's identity information is obtained through a multi-frequency radio frequency identification system to generate personalized adaptation data to preliminarily configure the shape of the smart seat;

[0051] Preferably, the intelligent visual sensor includes multiple camera modules, and the camera modules are configured to capture the user's facial expression features, posture features and body features from multiple perspectives to generate personalized adaptation data, wherein the personalized adaptation data is used to adjust the initial form of the smart seat to make it more closely fit the user's body contour.

[0052] In the present invention, the intelligent visual sensor realizes the function of capturing the user's facial expression features, posture features and body shape features from multiple perspectives through the configuration of multiple camera modules. The configuration principle of this multi-camera module is based on stereo vision technology, and image data is obtained from different angles through multiple cameras to reconstruct the user's three-dimensional morphological information. After stereo matching and geometric correction, the image captured by each camera can generate a more accurate depth map and user profile data, thereby forming personalized adaptation data.

[0053] In stereo vision, multiple camera modules simultaneously capture user images from different angles, and the user's three-dimensional features can be calculated using these image data. Specifically, the camera module collects the user's facial expression features, and captures the user's emotions and postures by analyzing the changes in expression; at the same time, the user's posture features and body shape features are obtained to determine the user's sitting posture and body shape. These feature data are processed by computer vision algorithms to generate personalized adaptation data, which is used to guide the initial shape configuration of the smart seat, so that the seat can dynamically adapt to the user's body shape and posture changes to ensure comfort.

[0054] This multi-view capture technology has a significant effect, improving the accuracy and flexibility of seat adaptation. Through personalized adaptation data, the smart seat can automatically adjust its shape the moment the user sits down, so that the seat back and cushion fit closely to the user's body, thereby improving the comfort and support of the seat. Specifically, personalized adaptation data allows smart material components to dynamically adjust according to the user's body contours, reduce local pressure concentration, and avoid physical discomfort caused by long-term improper sitting posture. In addition, facial expression features captured by the camera module can be used to assess the user's fatigue state and emotional changes, further enhancing the user experience. For example, when it is detected that the user is showing fatigue, the seat can automatically adjust to a more comfortable tilt angle.

[0055] In actual implementation, the smart vision sensor can be composed of multiple high-resolution cameras installed in different positions of the seat, such as the top, sides or front. Through computer vision software, the system processes the image data received from each camera in real time and extracts key features using deep learning algorithms. These algorithms include but are not limited to convolutional neural networks (CNN) for facial feature recognition and 3D reconstruction technology for posture analysis. Through such implementation, the smart seat can adjust according to the user's immediate state, which not only improves comfort, but also improves the intelligence and interactivity of the seat.

[0056] Preferably, the multi-frequency RFID system obtains the user's identity information by detecting the RFID tag worn by the user. The multi-frequency RFID system is used to monitor and update the changes of the user's identity information in real time, generate dynamic personalized adaptation data, and continuously adjust the configuration parameters of the smart seat to match the user's current identity.

[0057] In the present invention, a multi-frequency radio frequency identification system (RFID) obtains the user's identity information by detecting the radio frequency tag worn by the user. This system plays a key role in the present invention, and can monitor and update the user's identity information in real time, thereby generating dynamic personalized adaptation data for continuous adjustment of the configuration parameters of the smart seat to match the user's current identity.

[0058] The multi-frequency RFID system consists of a set of RFID readers and RFID tags worn by users. Each user's RFID tag contains unique identification information. When the user approaches the seat, the RFID reader communicates with the tag through radio waves to identify and extract the user's identity information. The system supports multi-band communication and can identify at different frequencies to ensure reliability and accuracy in different environments. Once the user's identity information is identified, the system can access the user's personalized configuration file, including body shape, preferred sitting posture and historical adjustment records.

[0059] Using the RFID system, smart seats can adjust their configuration parameters in real time to suit the user's personalized needs. This automated identification and adaptation process greatly improves the user experience and the intelligence of the seat. For example, the seat can automatically adjust the support strength of the back and seat cushion according to the user's body shape, as well as adjust the seat's tilt angle and height based on historical usage data. In addition, when the user switches to a different seat, the RFID system can quickly identify the user and apply his or her personalized configuration to the new seat, so that each use can achieve optimal comfort and support.

[0060] In specific implementations, the RFID tag can be a bracelet worn by the user or embedded in the user's personal belongings (such as a keychain). The RFID reader is installed near the backrest or armrest of the seat to ensure that it can effectively cover the area when the user approaches. When the user enters the detection range of the seat, the RFID system will automatically identify and read the user's identity information, and then call the user's personalized adaptation data in the seat control system. The seat control system adjusts the various parameters of the seat in real time based on this data, including the inclination of the seat cushion, the height of the armrest and the strength of the back support, to ensure that the user can immediately obtain their preferred seat settings.

[0061] like Figure 3As shown, according to the personalized adaptation data, the user's back pressure data is collected in real time using smart material technology and nanosensor networks, pressure feature data is generated through a multi-scale modeling algorithm, and pressure adjustment data is generated using a recursive fuzzy control algorithm to dynamically adjust the morphology of the smart material;

[0062] In the present invention, the system uses intelligent material technology and nanosensor networks to collect real-time pressure data on the user's back based on personalized adaptation data. Subsequently, pressure feature data is generated through a multi-scale modeling algorithm, and pressure adjustment data is generated using a recursive fuzzy control algorithm to dynamically adjust the shape of the intelligent material, thereby achieving a close fit between the seat and the human back.

[0063] The core of the system is the combination of smart material technology and nanosensing network. Smart materials such as shape memory alloys or liquid crystal polymers can change shape and stiffness according to external input signals. The nanosensing network consists of a series of micro sensors distributed in the back and cushion of the seat to monitor the user's back pressure distribution in real time. These sensors can detect tiny pressure changes and generate high-resolution real-time pressure data. Through multi-scale modeling algorithms, these real-time pressure data are decomposed into components of different frequencies and spatial resolutions to generate pressure feature data to describe the specific pressure distribution characteristics of the user's back.

[0064] The application of a recursive fuzzy control algorithm enables the system to generate pressure adjustment data based on pressure signature data. The algorithm uses fuzzy logic to handle nonlinearity and uncertainty, compares the user's dynamic pressure distribution with the ideal pressure distribution, and calculates the required morphological adjustment path. Through recursive feedback adjustment, the smart material changes its morphology based on the generated pressure adjustment data, dynamically adapting to the user's back shape and sitting posture.

[0065] The effect of this real-time adjustment technology is that it can provide a highly personalized seating experience, improving user comfort and support. The dynamic adjustment ability of smart materials ensures that the seat can be adjusted at any time according to the user's body shape and sitting posture changes, avoiding physical discomfort and fatigue caused by improper sitting posture for a long time. The combination of nanosensor networks and multi-scale modeling algorithms provides accurate pressure distribution analysis, allowing the seat to provide uniform support in any situation, improve blood circulation, and reduce muscle fatigue.

[0066] In practice, the nanosensors built into the seat are arranged in a matrix, covering the entire back and cushion area. When the user sits down, the sensors immediately start collecting pressure data and perform multi-scale modeling analysis through the built-in processor. The real-time pressure characteristic data is transmitted to the control module via wireless signals, which applies a recursive fuzzy control algorithm to generate an adjustment signal and send it to the smart material. After receiving the signal, the smart material immediately changes its form to achieve adjustment, such as changing its flexibility or shape, to ensure that every part of the user's back can get proper support and pressure distribution.

[0067] Preferably, the multi-scale modeling algorithm is used to analyze the pressure data of the user's back to generate pressure characteristic data, wherein the pressure characteristic data is calculated by decomposing the real-time pressure data into multiple scales and performing weighted summation, and the expression is as follows:

[0068]

[0069] in, Indicates at time Time, location Pressure characteristic data at is the weight for each scale; is the spatial basis function; is the time basis function; is the scale quantity.

[0070] The multi-scale modeling algorithm is used to analyze the pressure data of the user's back and generate pressure feature data. This process is achieved by decomposing the real-time pressure data into multiple scales and weighted summing them, aiming to provide a more detailed and accurate description of the pressure distribution to help the smart seat better adapt to the user's back shape.

[0071] The multiscale modeling algorithm is a mathematical tool for processing complex data sets by breaking down the data into different scales to capture details and overall trends. Specifically, the algorithm breaks down the pressure data using basis functions in the spatial and temporal dimensions. Spatial basis functions It is used to represent the pressure distribution characteristics at different spatial positions (i.e., seat surface), while the time basis function Used to capture the dynamic characteristics of pressure changes. Pressure characteristic data is generated by weighted summation of each scale. It reflects the user's stress level at a specific time and location.

[0072] Each part of the formula has a clear physical meaning: Represents the weight coefficient of each scale, which determines the influence of different scales on the pressure characteristic data. is a spatial decomposition function that provides details of the spatial distribution of pressure. It is a time-decomposed function that reveals the change of pressure over time. In this way, the algorithm can extract the most representative feature information from complex pressure data to help smart seats achieve more precise adjustment.

[0073] The multi-scale modeling algorithm significantly improves the adaptability and comfort of the smart seat. By analyzing the pressure distribution on the user's back in detail, the system can identify possible areas of pressure concentration and make corresponding adjustments to avoid discomfort caused by local excessive pressure. At the same time, this detailed pressure characteristic data analysis provides a reliable basis for the dynamic adjustment of the seat, making the morphological changes of smart materials more targeted and efficient. In addition, multi-scale analysis can respond to the user's posture changes in real time, providing continuously optimized support and comfort.

[0074] In the specific implementation, the nanosensor network collects pressure data from the seat surface in real time and transmits it to the processing unit. The processing unit applies a multi-scale modeling algorithm to decompose the data into different spatial scales and temporal frequency bands. For example, the spatial basis function can be a local function based on the seat grid structure, while the temporal basis function may be based on the user's breathing and movement cycle. After the data at each scale is weighted and summed, the generated pressure characteristic data is used to adjust the shape of the smart material. Based on the guidance of this data, the smart material adjusts the support strength and contact surface to optimize the comfort of the seat in different areas.

[0075] Preferably, the recursive fuzzy control algorithm generates pressure adjustment data based on the pressure characteristic data, and the pressure adjustment data is used to dynamically adjust the shape of the smart material. The recursive fuzzy control algorithm calculates an adjustment path by recursively evaluating the difference between the pressure characteristic data and the target pressure distribution.

[0076] In the present invention, the recursive fuzzy control algorithm generates pressure adjustment data based on pressure characteristic data to achieve dynamic adjustment of the shape of the smart material. The algorithm calculates the adjustment path by recursively evaluating the difference between the pressure characteristic data and the target pressure distribution, thereby ensuring the best fit between the seat and the user's back.

[0077] The recursive fuzzy control algorithm combines the advantages of fuzzy logic and recursive calculation, and is suitable for processing complex nonlinear systems. In the present invention, the algorithm first obtains the pressure characteristic data generated by the multi-scale modeling algorithm, which reflects the pressure distribution of the user's current back contact surface with the seat. Then, the algorithm compares the actual pressure distribution with the predefined target pressure distribution. The target pressure distribution can be pre-set by the user's comfort preference or medical needs.

[0078] Through fuzzy logic, the algorithm is able to handle imprecise and ambiguous input data, converting pressure distribution differences into fuzzy rules. These rules are used to evaluate the difference between the current pressure distribution and the ideal distribution. The recursive process allows the algorithm to repeatedly calculate, continuously update and optimize the adjustment path to narrow the gap between the current state and the target state. Ultimately, the pressure adjustment data generated by the algorithm is used to guide the dynamic shape adjustment of the smart material to make it more consistent with the user's back curve.

[0079] The effect of this control method is to improve the responsiveness and adaptability of the seat, allowing the seat to automatically adjust its support and shape according to the user's body shape and posture changes. The recursive fuzzy control algorithm provides a smoother and more precise adjustment path, avoiding the discomfort caused by rapid or excessive adjustments. At the same time, the algorithm can adapt to the user's dynamic changes to achieve a continuous comfort experience. For example, when the user's sitting posture is slightly adjusted, the seat can immediately detect and adjust to maintain the best support and pressure distribution on the back.

[0080] In specific implementation, the control unit of the smart seat continuously receives and analyzes pressure signature data from the nanosensor network. The recursive fuzzy control algorithm runs through the built-in microprocessor and calculates the difference between the current pressure distribution and the target distribution in real time. Assuming that the user's target pressure distribution is uniform, the algorithm will detect any area with excessive local pressure and calculate the adjustment path that needs to reduce the pressure in that area. Subsequently, the smart material undergoes morphological changes based on the pressure adjustment data, such as reducing pressure concentration by changing the stiffness or shape of the material. This dynamic adjustment process is continuous, ensuring that the user is always in the best state of comfort.

[0081] like Figure 4 As shown in the figure, the user's sitting posture changes and micro-movements are collected to generate behavioral feature data, and behavioral pattern data is generated through time series analysis; the posture optimization parameters are generated using an adaptive neural network model, and the posture of the smart seat is adjusted in real time in combination with a multimodal feedback system;

[0082] In the present invention, the system generates behavioral feature data by collecting the user's sitting posture changes and micro-movements, and generates behavioral pattern data through time series analysis. This process combines an adaptive neural network model and a multimodal feedback system to adjust the posture of the smart seat in real time to improve the user's comfort and support.

[0083] Behavioral characteristic data is collected through a sensor array, which can monitor the user's sitting posture changes, subtle movements, and biological signals (such as electromyographic signals) in real time. These data constitute the dynamic characteristics of the user's interaction with the chair. In order to capture these dynamic changes, the system uses time series analysis methods to process and analyze the data and extract the user's behavioral pattern data. These pattern data can reflect the user's sitting habits, change frequency and tendency, and provide a basis for subsequent posture optimization.

[0084] The adaptive neural network model is one of the core algorithms of the system, which is used to process behavioral pattern data. The neural network generates posture optimization parameters by learning the long-term behavioral characteristics and real-time dynamic changes of users. These parameters indicate how the seat should adjust its shape to adapt to the user's immediate needs. The multimodal feedback system provides adjustment suggestions to users in real time by combining visual, tactile and audio feedback, and receives user feedback for further optimization.

[0085] The system achieves precise response and adaptation to user behavior by dynamically adjusting the posture of the smart seat. Since each user has different sitting postures and usage habits, it is difficult for traditional seats to provide personalized support, and the real-time adjustment function of the smart seat can significantly improve comfort and reduce fatigue and discomfort. For example, the system can detect that the user tends to lean forward or backward, and automatically adjust the inclination of the seat to provide better support. This dynamic responsiveness can also reduce the risk of long-term use of users due to poor sitting posture, such as lumbar pain or muscle fatigue.

[0086] In actual applications, smart seats are equipped with a variety of sensors, including accelerometers, gyroscopes, and pressure sensors, to collect user behavior characteristic data. These sensors are installed in key locations of the seat, such as the seat cushion, backrest, and armrests, to obtain comprehensive user behavior information. The collected data is input into the adaptive neural network model for processing after time series analysis. The model outputs posture optimization parameters to instruct the seat to make necessary adjustments, such as changing the tilt angle of the backrest or the softness of the seat cushion. The multimodal feedback system interacts with the user through the vibration of the seat, LED indicators, and voice prompts to improve the user's acceptance and satisfaction with the seat adjustment.

[0087] Preferably, the behavior characteristic data includes the user's sitting posture change amplitude, micro-movement frequency and electromyographic signal change, and the behavior characteristic data is used to analyze the user's behavior characteristics to identify the user's long-term posture pattern and usage habits.

[0088] In the present invention, behavioral feature data is an important basis for analyzing user behavioral features, mainly including the amplitude of changes in user sitting posture, frequency of micro-movements, and changes in electromyographic signals. These data are used to identify the user's long-term posture patterns and usage habits, thereby optimizing the seat adjustment strategy and improving the user's comfort and health.

[0089] Behavioral characteristic data is collected through a series of sensors, including accelerometers, gyroscopes, and electromyography sensors. Accelerometers and gyroscopes are used to detect the amplitude of changes in the user's sitting posture and the frequency of micro-movements. The amplitude of changes in sitting posture refers to the obvious posture adjustments made by the user in the chair, such as leaning forward, leaning back, and sideways. The frequency of micro-movements refers to the subtle movements made by the user when using the chair, such as slight twisting or posture adjustments. Changes in electromyography signals are detected by electromyography sensors, reflecting changes in the user's muscle activity, which can reveal the user's relaxed or tense state in the chair.

[0090] By analyzing these behavioral feature data, the system is able to identify users' long-term posture patterns and usage habits. For example, some users may tend to maintain a fixed posture while working, while other users may frequently adjust their sitting posture for comfort. This pattern recognition process not only helps provide users with personalized chair adjustment solutions, but also helps identify bad sitting habits, thereby making effective posture correction suggestions.

[0091] The effect of this behavioral analysis is that the smart chair can provide a highly personalized comfort experience and help users develop healthy sitting habits. By continuously monitoring the user's behavioral characteristic data, the chair can adjust the support strength and angle in real time to adapt to the user's dynamic needs and prevent muscle tension and fatigue caused by long-term poor sitting posture. Long-term pattern recognition can also be used to create user profiles to help users understand their own sitting habits and provide improvement suggestions.

[0092] In specific implementation, the sensor arrays equipped in the smart seat are distributed in key positions of the seat back and seat cushion to collect user behavior characteristic data in real time. Accelerometers and gyroscopes can record the amplitude of changes in the user's sitting posture and the frequency of micro-movements, while electromyographic sensors measure the electromyographic signals of the user's back and legs. The collected data is transmitted to the data processing unit via a wireless network for pattern recognition and analysis. Based on the identified long-term posture patterns, the system automatically generates adjustment strategies. For example, if the user often leans forward, the seat can automatically increase lumbar support to reduce back pressure.

[0093] Preferably, the time series analysis analyzes the behavior feature data by short-term Fourier transform to generate behavior pattern data, and the behavior pattern data captures the dynamic characteristics and periodic characteristics of the user's posture changes.

[0094] In the present invention, time series analysis analyzes the behavior feature data through short-term Fourier transform (STFT) to generate behavior pattern data. The purpose of this process is to capture the dynamic characteristics and periodic characteristics of user posture changes, thereby providing a more intelligent and accurate basis for adjusting the seat.

[0095] Short-term Fourier transform (STFT) is a commonly used signal processing technology used to analyze the changes in the frequency components of non-stationary signals over time. For smart seat systems, the user's sitting posture and micro-movement data are non-stationary signals because these data change over time. By performing STFT analysis on these data, the system can simultaneously observe the spectral distribution and time changes of the signal in the time-frequency domain, thereby identifying the dynamic characteristics and periodic characteristics of the user's posture changes.

[0096] Specifically, the system divides the behavior feature data into several short-term windows and performs Fourier transform on each window to obtain the frequency information of the signal. Then, by combining the frequency information of all windows, complete behavior pattern data is generated. These pattern data can reveal the frequency and amplitude of changes in the user's sitting posture, as well as specific periodic behaviors (such as timed adjustments or relaxation movements), thus providing a reference for the intelligent adjustment of the seat.

[0097] Through time series analysis and STFT, the smart chair can more accurately understand the user's sitting habits and change patterns, thereby providing more personalized support and comfort. Identifying the user's dynamic characteristics and periodic characteristics allows the chair to predict the user's adjustment needs in advance and make proactive adjustments. For example, if the user regularly adjusts his posture at the end of each hour, the chair can detect this pattern and make fine adjustments in advance on the hour to provide more comfortable support. Such functions not only improve the user experience, but also help users develop healthier sitting habits.

[0098] In actual applications, sensors in the smart seat system collect user posture change data in real time, and this data is input into the processing unit for STFT analysis. Assuming a user regularly adjusts his sitting posture, this behavior will show up as a specific frequency peak in the STFT analysis. By analyzing these frequency components, the system is able to identify the user's behavior pattern and generate behavior pattern data. This data is then used to adjust the seat's tilt angle and support force to better meet the user's dynamic needs.

[0099] Preferably, the adaptive neural network model generates posture optimization parameters by analyzing the behavior pattern data, and the calculation process is implemented by the following expression:

[0100]

[0101] in, Parameters for posture optimization, indicating how the seat should be adjusted for comfort and support; Represents a neural network model; Behavioral pattern data; is the time step, which is used to guide the real-time adjustment of the intelligent seat posture. This structure enables the neural network to dynamically adapt to the user's behavior changes and provide personalized seat adjustment suggestions.

[0102] In the present invention, the adaptive neural network model generates posture optimization parameters by analyzing the behavior pattern data. This process is the core of the system to achieve intelligent posture adjustment. The neural network model uses the user's behavior pattern data to generate posture optimization parameters. and time step To calculate the posture optimization parameters ,This parameter is directly used to guide the real-time posture adjustment of the smart seat.

[0103] The adaptive neural network model is a computational model capable of complex nonlinear mapping. Its design is inspired by the biological nervous system and is suitable for analyzing and predicting complex data patterns. In the present invention, the neural network model uses behavioral pattern data as input to learn the user's posture changes and adjustment habits. The behavioral pattern data includes the user's sitting posture changes, micro-movement frequency, and periodic characteristics, etc., which are obtained through time series analysis. The neural network model uses this data for training and adaptation to generate posture optimization parameters that can reflect the user's current needs and future adjustment trends.

[0104] Through the adaptive neural network model, the smart chair can respond to the user's dynamic needs and changes in real time, thereby achieving a personalized comfort experience. The generated posture optimization parameters can not only adjust the chair according to the current behavior pattern, but also predict the user's future adjustment needs. For example, if it is detected that the user tends to lean back and relax after working for a long time, the seat can be pre-adjusted to a more comfortable tilt angle. This not only improves the user's comfort, but also effectively reduces health problems caused by poor sitting posture.

[0105] In actual applications, the sensors built into the seat continuously collect the user's behavioral characteristic data, analyze it through the data processing unit, and generate behavioral pattern data. The adaptive neural network receives this data and uses historical behavioral data and current real-time input for training and prediction. The generated posture optimization parameters are passed to the seat's control system to guide it to automatically adjust the shape of the backrest and seat cushion when the user sits down to provide optimal support and comfort.

[0106] For example, if the user's behavioral pattern data shows that they regularly adjust their sitting posture to relieve fatigue after working for a long time, the neural network will learn this pattern and actively adjust the seat angle at the appropriate moment to provide more ergonomic support. This process is dynamic and continuous, ensuring that the smart chair can always provide support and comfort that meets the user's needs.

[0107] Integrate personalized adaptation data, pressure adjustment data and posture optimization parameters, generate collaborative optimization data through collaborative analysis of multi-agent systems, and perform self-adjustment and optimization evaluation of the intelligent seat system.

[0108] Preferably, the collaborative analysis of the multi-agent system involves information exchange and joint decision-making among multiple intelligent agents. The intelligent agents are responsible for analyzing different data dimensions in the standardized data set and generating collaborative optimization data through comprehensive analysis, thereby realizing self-adjustment and optimization evaluation of the intelligent seat system.

[0109] In the present invention, by integrating personalized adaptation data, pressure adjustment data and posture optimization parameters, the collaborative analysis of the multi-agent system is used to generate collaborative optimization data, so as to realize self-adjustment and optimization evaluation of the intelligent seat system. This process is an important step to ensure that the seat can dynamically adapt to user needs.

[0110] A multi-agent system is a distributed computing framework consisting of multiple intelligent agents, each of which is responsible for specific data analysis tasks and decision making. In the present invention, the multi-agent system is designed to process multiple data sources from smart seats, including personalized adaptation data, pressure adjustment data, and posture optimization parameters. Each agent focuses on a specific data dimension, for example, some agents focus on analyzing the user's body shape data, some focus on real-time pressure changes, and some are responsible for the evaluation of posture optimization strategies.

[0111] Through information exchange and joint decision-making, each intelligent agent can share its own analysis results and jointly generate a comprehensive collaborative optimization data set. This collaborative optimization data not only contains the independent analysis results of each data source, but also includes the relationship and comprehensive effect between them. This data set is used to guide the overall adjustment strategy of the intelligent seat, ensuring that all adjustment measures are based on a comprehensive understanding of the user's status.

[0112] This collaborative analysis of the multi-agent system greatly improves the responsiveness and adjustment accuracy of the smart seat. By considering multiple data dimensions and user needs at the same time, the smart seat can better adapt to the user's personalized needs and provide more comfortable and ergonomic support. For example, when the personalized adaptation data shows that the user has special support needs, the seat can adjust its pressure distribution and posture support through comprehensive analysis to meet these needs.

[0113] At the same time, through continuous self-adjustment and optimization evaluation, the seat can adapt to changes in different users and usage scenarios to achieve lasting comfort and health support. The optimization evaluation process also helps identify and correct possible adjustment deviations to ensure that the system can continue to maintain its optimal state.

[0114] In actual applications, each intelligent agent in the smart seat is deployed in the central control unit. Personalized adaptation data is provided by the user's recognition system, pressure adjustment data is collected in real time by the sensor network, and posture optimization parameters are generated by the neural network model. These data are standardized and input into the multi-agent system.

[0115] Each intelligent agent uses machine learning and data analysis techniques to process the data dimension it is responsible for. Through a central coordinator, all agents share analysis results and make joint decisions based on the overall strategy. For example, when a user needs better rest support after a long work day, multiple agents work together to adjust the seat's tilt and pressure distribution, while adjusting the support strength of the back and legs to achieve comprehensive posture optimization.

[0116] like Figure 5 As shown, a system for implementing the method for interacting with the intelligent seat that fits the back of a human body comprises:

[0117] The intelligent visual sensor module is configured to collect the user's facial expression features, posture features and body shape features, and generate visual feature data to provide the user's personalized adaptation data for the preliminary morphological configuration of the intelligent seat; the module captures the user's facial expression features, posture features and body shape features in real time through a camera array installed around the seat. In principle, computer vision technology is used to process these image data to extract the user's physiological characteristics and current posture. The generated visual feature data is used to construct the user's personalized adaptation data, which is used to preliminarily configure the morphology of the intelligent seat. In this way, the seat can immediately adapt to the user's body shape and sitting posture, providing more fitting support. For example, when the user sits down, the seat automatically adjusts the shape of the backrest to fit the user's back curve, thereby improving comfort and support.

[0118] The multi-frequency radio frequency identification module is used to obtain the user's identity information and generate personalized adaptation data to ensure that the configuration of the smart seat matches the user's identity; this module is used to identify the user's identity through a multi-frequency RFID tag and a reader. The RFID tag carried by each user contains unique identity information. When the user enters the seat area, the RFID system identifies and reads the information to generate personalized adaptation data. This data ensures that the configuration of the seat can match the user's personal habits and preferences. In effect, the seat can automatically adjust settings based on the user's historical data to provide a personalized experience, such as adjusting the seat height and tilt angle to meet the user's comfort needs.

[0119] Smart material components, including flexible materials configured to dynamically respond to pressure adjustment data to adjust the shape of the smart seat to achieve dynamic fit with the user's back; the smart material components are composed of flexible materials that can dynamically respond to pressure adjustment data. These materials can change their physical properties (such as stiffness or shape) to adapt to the dynamic fit requirements of the user's back. Through real-time adjustments, the smart material achieves a perfect fit with the user's back, thereby improving the comfort and support of the seat. For example, when the user leans back, the seat back will deform to provide more support.

[0120] The nanosensor network, embedded in the smart seat, is used to collect real-time pressure data on the user's back and generate real-time pressure data for processing by the multi-scale modeling algorithm; the nanosensor network consists of tiny sensors embedded in the seat to collect real-time pressure data on the user's back. These sensors can detect and record changes in pressure distribution on the seat surface, generating real-time pressure data for subsequent processing. In effect, this high-precision data collection can help the seat better understand the details of the user's contact with the seat, thereby optimizing the seat's shape and support.

[0121] The multi-scale modeling processing unit is used to process the real-time pressure data generated by the nanosensing network and generate pressure feature data through the multi-scale modeling algorithm; the processing unit is responsible for analyzing the real-time pressure data generated by the nanosensing network. Through the multi-scale modeling algorithm, the data is decomposed into features of different scales to capture the detailed pressure distribution of the user's back. These pressure feature data are used to guide the shape adjustment of the seat, ensure the uniformity of pressure distribution, and reduce user discomfort.

[0122] The recursive fuzzy controller is used to receive pressure characteristic data and generate pressure adjustment data through a recursive fuzzy control algorithm to guide the dynamic adjustment of the smart material component; the recursive fuzzy controller uses fuzzy logic to process complex pressure characteristic data and generate pressure adjustment data. The algorithm continuously adjusts the support force of the seat to adapt to the dynamic needs of the user through recursive calculation. This dynamic adjustment enables the seat to respond to the user's posture changes in a short time, improving the user experience.

[0123] The behavior analysis unit is used to collect the user's sitting posture changes and micro-movements, generate behavior feature data, and generate behavior pattern data through time series analysis; the behavior analysis unit collects the user's sitting posture changes and micro-movements through built-in sensors to generate behavior feature data. Time series analysis is used to process these data to generate behavior pattern data. These data reveal the user's sitting habits and dynamic characteristics, providing a basis for seat adjustment.

[0124] The adaptive neural network model is used to process the behavior pattern data and generate posture optimization parameters to adjust the posture of the smart seat; the adaptive neural network model uses the behavior pattern data to generate posture optimization parameters for real-time adjustment of the seat posture. The model optimizes the support and comfort of the seat by learning the user's dynamic behavior pattern.

[0125] The multimodal feedback system is used to provide real-time user feedback to help adjust the posture of the smart seat; the multimodal feedback system provides user feedback through multiple means such as vision, hearing and touch to help users adjust the posture. The feedback mechanism of the system ensures that users can receive seat adjustment suggestions in a timely manner, improving user satisfaction.

[0126] The multi-agent system is used to integrate personalized adaptation data, pressure adjustment data and posture optimization parameters, generate collaborative optimization data through collaborative analysis, and perform self-adjustment and optimization evaluation of the intelligent seat system. The multi-agent system integrates personalized adaptation data, pressure adjustment data and posture optimization parameters, generates collaborative optimization data through collaborative analysis. The system performs self-adjustment and optimization evaluation to achieve dynamic adaptation and personalized experience of the intelligent seat system. Through the collaboration of multiple intelligent agents, the system can provide optimal support and comfort in different situations.

[0127] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware.

[0128] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An interactive method for an intelligent seat that fits the back of a human body, characterized in that: The following steps are involved: The user's facial expression features, posture features and body features are collected through an intelligent visual sensor, and the user's identity information is obtained through a multi-frequency radio frequency identification system to generate personalized adaptation data to preliminarily configure the shape of the smart seat; the intelligent visual sensor includes a plurality of camera modules, and the camera modules are configured to capture the user's facial expression features, posture features and body features from multiple perspectives to generate personalized adaptation data, wherein the personalized adaptation data is used to adjust the initial shape of the smart seat to make it more closely fit the user's body contour; the multi-frequency radio frequency identification system obtains the user's identity information by detecting the radio frequency tag worn by the user, and the multi-frequency radio frequency identification system is used to monitor and update the changes in the user's identity information in real time, and generate dynamic personalized adaptation data, so as to continuously adjust the configuration parameters of the smart seat to match the user's current identity; According to the personalized adaptation data, the user's back pressure data is collected in real time using smart material technology and nanosensor network, pressure characteristic data is generated through a multi-scale modeling algorithm, and pressure adjustment data is generated using a recursive fuzzy control algorithm to dynamically adjust the shape of the smart material. The dynamic adjustment includes: reducing pressure concentration by changing the stiffness or shape of the material; the multi-scale modeling algorithm is used to analyze the user's back pressure data and generate pressure characteristic data, wherein the pressure characteristic data is calculated by decomposing the real-time pressure data into multiple scales and weighted summing them, and the expression is as follows: in, Indicates at time Time, location Pressure characteristic data at is the weight for each scale; is the spatial basis function; is the time basis function; is the scale quantity; The recursive fuzzy control algorithm generates pressure adjustment data based on the pressure characteristic data, and the pressure adjustment data is used to dynamically adjust the shape of the smart material. By recursively evaluating the difference between the pressure characteristic data and the target pressure distribution, the recursive fuzzy control algorithm calculates the adjustment path The user's sitting posture changes and micro-movements are collected to generate behavioral feature data, and behavioral pattern data is generated through time series analysis; the posture optimization parameters are generated using an adaptive neural network model, and the posture of the smart seat is adjusted in real time in combination with a multimodal feedback system; the behavioral feature data includes the user's sitting posture change amplitude, micro-movement frequency, and electromyographic signal changes, and the behavioral feature data is used to analyze the user's behavioral characteristics to identify the user's long-term posture pattern and usage habits; the time series analysis analyzes the behavioral feature data through short-term Fourier transform to generate behavioral pattern data, and the behavioral pattern data captures the dynamic characteristics and periodic characteristics of the user's posture changes; the adaptive neural network model generates posture optimization parameters by analyzing the behavioral pattern data, and the calculation process is implemented by the following expression: in, Optimize parameters for posture; Represents a neural network model; Behavioral pattern data; is the time step, which is used to guide the real-time adjustment of the intelligent seat posture; Integrate personalized adaptation data, pressure adjustment data and posture optimization parameters, generate collaborative optimization data through collaborative analysis of multi-agent systems, and perform self-adjustment and optimization evaluation of the intelligent seat system.

2. The method for interacting with a smart chair that fits the back of a human body according to claim 1, characterized in that: The collaborative analysis of the multi-agent system involves information exchange and joint decision-making among multiple intelligent agents. The intelligent agents are responsible for analyzing different data dimensions in the standardized data set and generating collaborative optimization data through comprehensive analysis, thereby realizing self-adjustment and optimization evaluation of the intelligent seat system.

3. A system for implementing the intelligent seat interaction method for fitting with the back of a human body as described in any one of claims 1-2, characterized in that: include: An intelligent visual sensor module, configured to collect facial expression features, posture features and body shape features of a user, and generate visual feature data to provide user personalized adaptation data for preliminary configuration of the intelligent seat; Multi-frequency radio frequency identification module, used to obtain the user's identity information and generate personalized adaptation data to ensure that the configuration of the smart seat matches the user's identity; a smart material assembly including a flexible material configured to dynamically respond to pressure modulation data to adjust the form of the smart seat to achieve a dynamic fit to a user's back; Nanosensor network, embedded in the smart chair, is used to collect real-time pressure data on the user's back and generate real-time pressure data for processing by multi-scale modeling algorithms; A multi-scale modeling processing unit, used to process the real-time pressure data generated by the nanosensor network and generate pressure characteristic data through a multi-scale modeling algorithm; a recursive fuzzy controller for receiving pressure characteristic data and generating pressure adjustment data through a recursive fuzzy control algorithm to guide dynamic adjustment of the smart material component; The behavior analysis unit is used to collect the user's sitting posture changes and micro-movements, generate behavior feature data, and generate behavior pattern data through time series analysis; An adaptive neural network model for processing behavioral pattern data and generating posture optimization parameters to adjust the posture of the smart seat; Multimodal feedback system, used to provide real-time user feedback to help adjust the posture optimization of smart seats; A multi-agent system is used to integrate personalized adaptation data, pressure regulation data, and posture optimization parameters, generate collaborative optimization data through collaborative analysis, and perform self-adjustment and optimization evaluation of the intelligent seat system.

Images