A sitting posture monitoring control system of an intelligent learning desk and chair

Abstract

The application discloses a sitting posture monitoring control system of an intelligent learning desk and chair, and relates to the technical field of smart home and ergonomics equipment. The system comprises a sitting posture data acquisition module, which is used for acquiring multi-dimensional original data of a user's body posture in real time; a sitting posture state analysis module, which is used for fusion calculation and feature extraction based on the multi-dimensional original data collected by the sitting posture data acquisition module, accurate identification of a current sitting posture state and judgment of whether the current sitting posture state deviates from a health benchmark; a control strategy generation module, which is used for generating optimal correction control instructions according to the type and degree of sitting posture deviation, in combination with historical sitting posture data and personalized parameters; and an actuator driving module, which is used for accurately driving corresponding mechanical execution units of the desk and chair to adjust according to the control instructions. The application evolves from passive correction to active prevention through multi-sensor fusion perception, real-time feature extraction and personalized control strategy generation, and comprehensively monitors and dynamically corrects the sitting posture state of a user.

Description

Technical Field

[0001] This invention relates to the field of smart home and ergonomic equipment technology, and in particular to a posture monitoring and control system for a smart study desk and chair. Background Technology

[0002] As the comfort and health of learning environments become important development directions in the fields of smart home and educational technology, smart study desks and chairs, as key products in this field, aim to provide users with a more personalized and healthy learning experience by integrating sensors and control systems.

[0003] Among them, posture monitoring and control is the core technology of intelligent learning desk and chair system. The goal is to detect the user's posture in real time and intervene and correct it in time when bad posture is detected, so as to prevent health problems such as myopia and scoliosis.

[0004] Existing technologies typically rely on a single type of sensor for posture detection, such as using only a pressure sensor to determine hip pressure distribution or a distance sensor to monitor the user's distance from the desktop. Such solutions struggle to comprehensively and accurately capture complex sitting postures and are prone to misjudgments due to slight user movements or clothing obscuring the view. Furthermore, existing systems often employ simplistic control strategies, merely issuing audible and visual alarms without the ability to coordinate with the desk and chair hardware, failing to adaptively correct posture issues based on the user's specific posture. In addition, these systems typically do not consider the accumulation and analysis of long-term posture data, making it difficult to provide personalized posture improvement suggestions. Summary of the Invention

[0005] The purpose of this invention is to provide a posture monitoring and control system for intelligent study desks and chairs, in order to solve the problems of existing intelligent study desk and chair posture monitoring systems, such as single monitoring dimensions, rigid control strategies, and lack of long-term personalized adaptability.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A posture monitoring and control system for an intelligent study desk and chair includes: a posture data acquisition module; a posture state analysis module; a control strategy generation module; and an actuator drive module. The posture data acquisition module is responsible for acquiring multi-dimensional raw data of the user's body posture in real time. The posture state analysis module performs fusion calculations and feature extraction based on the acquired multi-dimensional raw data to accurately identify the current posture state and determine whether it deviates from a health benchmark. The control strategy generation module generates optimal corrective control commands based on the type and degree of posture deviation, combined with historical posture data and personalized parameters. The actuator drive module then precisely drives the corresponding mechanical actuators of the desk and chair to adjust according to the control commands, thus forming a closed-loop control system encompassing monitoring, analysis, decision-making, and execution.

[0008] The posture data acquisition module further includes a pressure distribution sensing unit, a three-dimensional posture perception unit, and a distance detection unit. The pressure distribution sensing unit consists of a high-density pressure sensor array embedded in the seat surface. This array is uniformly distributed with a density of at least one sensing point per square centimeter, used to measure the pressure distribution map of the user's buttocks and back in real time. The three-dimensional posture perception unit is implemented by a miniature inertial measurement unit deployed at the top of the seat back. This miniature inertial measurement unit integrates a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer, continuously acquiring the tilt angle, angular velocity, and azimuth angle data of the backrest in three-dimensional space at a sampling frequency of 100 Hz. The distance detection unit includes two sets of infrared ranging sensors. One set is installed inside the midpoint of the front edge of the table to detect the horizontal distance between the user's sternum and the table, and the other set is installed below the table to detect the horizontal distance between the user's abdomen and the table.

[0009] The sitting posture analysis module further includes a data fusion calculation submodule, a sitting posture feature extraction submodule, and a health benchmark comparison submodule. The data fusion calculation submodule uses an extended Kalman filter algorithm to align the timestamps and unify the spatial coordinates of asynchronous heterogeneous data from the pressure distribution sensing unit, the 3D posture perception unit, and the distance detection unit, eliminating inherent random noise and temperature drift errors from each sensor, and outputting a 5-dimensional state vector containing the pressure center coordinates, trunk pitch angle, trunk tilt angle, chest-to-table distance, and abdomen-to-table distance. The sitting posture feature extraction submodule extracts three key feature parameters from the 5-dimensional state vector: the first parameter is the pressure distribution asymmetry, defined as the ratio of the absolute value of the difference between the pressure sums in the left and right halves to the total pressure; the second parameter is the trunk tilt angle, obtained by coordinate transformation of the pitch angle; and the third parameter is the chest-to-abdomen distance difference, calculated as the chest-to-table distance minus the abdomen-to-table distance. The health benchmark comparison submodule has a preset set of standard healthy sitting posture range values. These range values ​​are obtained through statistical analysis of a large number of standard sitting posture samples, specifically including: pressure distribution asymmetry less than or equal to 5%; trunk forward tilt angle between 85 degrees and 95 degrees; and chest-abdomen distance difference between 3 cm and 8 cm. The sitting posture analysis module compares the three feature parameters extracted in real time with the preset standard healthy sitting posture range values ​​one by one. When any feature parameter exceeds the corresponding standard range, the current sitting posture is determined to be poor, and a sitting posture deviation signal containing the specific deviation parameter identifier and the deviation amount is output.

[0010] The control strategy generation module further includes a strategy mapping submodule and a personalized correction submodule. The strategy mapping submodule stores a preset control strategy mapping table, which defines the correspondence between different types of posture deviation signals and preliminary control commands. Specifically, when the posture deviation signal indicates excessive pressure distribution asymmetry, the preliminary control command is to activate the seat leveling mechanism; when it indicates excessive torso tilt angle, the preliminary control command is to activate the backrest tilt angle adjustment mechanism and the desktop height adjustment mechanism; when it indicates excessive chest-abdomen distance difference, the preliminary control command is to activate the seat fore-aft position adjustment mechanism. The personalized correction submodule dynamically corrects the adjustment range of the preliminary control commands based on the user's historical posture data and personal physiological parameters stored in local non-volatile memory. The control strategy generation module calculates the average deviation of the same type of poor posture experienced by the user in the past 7 days, queries the user's pre-entered height and weight data, calculates a personalized adjustment coefficient for the user using a linear regression model, multiplies this coefficient by the baseline adjustment amount in the preliminary control command, and generates the final personalized control command.

[0011] The actuator drive module further includes an instruction parsing submodule and a motor drive submodule. The instruction parsing submodule receives personalized control instructions from the control strategy generation module and decodes these instructions into specific mechanism action commands. These mechanism action commands include a target mechanism identifier, target position coordinates, and motion speed parameters. The motor drive submodule generates corresponding pulse width modulation signals based on the mechanism action commands, driving the corresponding DC servo motor for precise movement. The actuator drive module controls a total of four actuators: a seat leveling mechanism that controls the left-right tilt angle of the seat surface via a motor, with an adjustment range of -3 degrees to +3 degrees; a backrest tilting mechanism that controls the backrest pitch angle via a motor, with an adjustment range of 90 degrees to 120 degrees; a desktop height adjustment mechanism that controls the vertical height of the desktop via a motor, with an adjustment range of 65 cm to 80 cm; and a seat fore-and-aft position adjustment mechanism that controls the forward-backward displacement of the seat in the horizontal direction via a motor, with an adjustment range of -10 cm to +10 cm. All motors are equipped with high-precision rotary encoders, forming a closed-loop position control system to ensure that the actuators can accurately reach the target position set by the command.

[0012] The posture analysis module also includes a posture trend prediction unit. This prediction unit uses a first-order autoregressive model to predict posture changes within the next 5 seconds based on a historical 5-dimensional state vector sequence over a preset time period; the preset time period is the most recent 30 sampling cycles. If the prediction result indicates that a certain characteristic parameter will exceed the healthy baseline range within the next 5 seconds, the posture analysis module will output a posture deviation warning signal in advance. This warning signal also triggers the control strategy generation module to generate corresponding preventative control commands, thereby upgrading the function from passive correction to active prevention.

[0013] The control strategy generation module also integrates a learning optimization unit. This unit continuously records the user's posture response data after each control command execution, specifically recording the percentage of time the user returns to a standard sitting posture within 5 minutes of execution. When the same user triggers the same type of poor posture for the third time in the same scenario, the learning optimization unit analyzes the execution effect data of the previous two control commands and uses gradient descent to fine-tune the parameters of the linear regression model in the personalized correction submodule. This allows the system to adaptively optimize the control strategy for the user, gradually improving correction efficiency.

[0014] Compared with the prior art, the beneficial technical effects of the present invention are as follows:

[0015] This invention fundamentally solves the technical contradictions of existing technologies, such as single-dimensional posture monitoring, rigid control strategies, and lack of long-term learning ability, by constructing a complete closed-loop control system that integrates multi-sensor data fusion, real-time sitting posture feature analysis, personalized control strategy generation, and precision actuator drive.

[0016] This invention employs multi-source data fusion calculation of pressure distribution, three-dimensional posture, and distance information to significantly improve the accuracy and robustness of sitting posture recognition, effectively avoiding misjudgment problems caused by the limitations of a single sensor.

[0017] This invention ensures the accuracy and adaptability of corrective measures by accurately comparing real-time sitting posture characteristics with preset health benchmarks and making personalized modifications to control commands based on historical data and individual parameters.

[0018] This invention introduces a sitting posture trend prediction and learning optimization mechanism, evolving from passive correction to proactive prevention. It continuously optimizes control strategies based on long-term user data to achieve personalized, adaptive sitting posture health management, realizing practical value in preventing myopia and spinal curvature abnormalities in adolescents. Attached Figure Description

[0019] Figure 1This is a schematic diagram of the overall technical solution architecture of the intelligent learning desk and chair posture monitoring and control system proposed in this invention;

[0020] Figure 2 This is a schematic diagram of the core principle framework of multi-sensor fusion perception and closed-loop control in this invention. Detailed Implementation

[0021] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely intended to explain the present invention and not to limit the present invention. For those skilled in the art, the present invention can be practiced without some of these specific details. The following description of the embodiments is merely to provide a better understanding of the present invention by illustrating examples of the invention.

[0022] Example 1

[0023] refer to Figure 1 This embodiment details the technical implementation of a posture monitoring and control system for an intelligent study desk and chair. The posture monitoring and control system constitutes a complete closed-loop control system. Its core objective is to achieve accurate monitoring and adaptive dynamic correction of the user's posture state through multi-source data fusion and intelligent decision-making. The posture monitoring and control system mainly includes four functional modules: a posture data acquisition module, a posture state analysis module, a control strategy generation module, and an actuator drive module. These four modules are connected sequentially to form a seamless data flow and control flow from data perception to physical execution.

[0024] The posture data acquisition module is the front end of a smart study desk and chair's posture monitoring and control system, responsible for sensing external physical quantities and acquiring multi-dimensional raw data on the user's body posture in real time. The posture data acquisition module specifically includes three functional units: a pressure distribution sensing unit; a three-dimensional posture perception unit; and a distance detection unit. The pressure distribution sensing unit consists of a high-density pressure sensor array embedded in the seat surface. The high-density pressure sensor array is uniformly distributed on the seat surface at a density of no less than one sensing point per square centimeter, ensuring that subtle changes in the pressure on the user's buttocks and back can be captured. Each pressure sensing point in the high-density pressure sensor array uses the piezoresistive sensing principle, and the output signal of the high-density pressure sensor array is a voltage value that is linearly related to the applied positive pressure. The voltage signal is digitized by a 24-bit precision analog-to-digital converter, with a sampling frequency set to 50 Hz. The digitized pressure data is organized into a two-dimensional matrix, with the row and column indices corresponding to the physical position coordinates of the sensors on the seat surface. The smart study desk and chair's posture monitoring and control system scans the entire matrix to calculate and update a complete pressure distribution map in real time. The three-dimensional attitude sensing unit is implemented by a miniature inertial measurement unit (INS) deployed at the top of the seat back. The INS utilizes system-in-package (SIP) technology to integrate a three-axis accelerometer, a three-axis gyroscope, and a three-axis magnetometer into a package measuring only 10 mm x 10 mm x 2 mm. The three-axis accelerometer measures the linear acceleration of the backrest in three orthogonal directions, with a range set to ±8 times the gravitational acceleration. The three-axis gyroscope measures the angular velocity of the backrest around three coordinate axes, with a range set to ±500 degrees per second. The three-axis magnetometer measures the ambient magnetic field strength, providing an absolute orientation reference for the posture monitoring and control system of the smart study desk and chair. The INS synchronously acquires 9-axis inertial data at a sampling frequency of 100 Hz and transmits the raw data stream to the posture analysis module via an integrated circuit bus interface. The distance detection unit includes two independent infrared ranging sensors. The first set of infrared range sensors is installed on the inner side of the midpoint of the front edge of the desktop. The optical axes of its infrared emitter and receiver are tilted downwards at a 15-degree angle to the desktop, ensuring that its detection beam is aimed at the user's xiphoid process. The second set of infrared range sensors is installed directly below the desktop, with its optical axis perpendicular to the desktop. It is used to detect the vertical distance between the user's abdomen and the desktop. Both sets of sensors use the phase comparison method for distance measurement, with a range of 5 cm to 50 cm, a resolution of 1 mm, and a data update frequency of 20 Hz. The distance output from the sensors is an analog voltage signal, which is converted into a digital value by a 12-bit analog-to-digital converter.

[0025] The posture analysis module receives raw data from the posture data acquisition module and performs fusion calculations and feature extraction on the raw data to accurately identify the current posture state. The posture analysis module comprises three core sub-modules: a data fusion calculation sub-module; a posture feature extraction sub-module; and a health benchmark comparison sub-module. The primary task of the data fusion calculation sub-module is to perform spatiotemporal alignment of asynchronous heterogeneous data from different sensor units. The posture monitoring and control system of the intelligent learning desk and chair uses a high-precision hardware clock to assign a unified timestamp to all incoming data. For data streams with different sampling frequencies, the data fusion calculation sub-module uses a linear interpolation algorithm to boost the lower frequency data to 100 Hz, keeping it synchronized with the inertial measurement unit data. Regarding the spatial coordinate system, the data fusion calculation sub-module defines a right-handed Cartesian coordinate system with the geometric center of the seat as the origin, the positive X-axis pointing directly forward as the positive Y-axis pointing to the user's left as the positive Z-axis, and the vertically upward direction as the positive Z-axis. All sensor measurements are transformed to this unified coordinate system. Subsequently, the data fusion calculation sub-module uses an extended Kalman filter algorithm to fuse and denoise the data. The filter's state variables are defined as a 5-dimensional vector, including the X and Y coordinates of the pressure center, trunk pitch angle, trunk tilt angle, distance from the chest to the table, and distance from the abdomen to the table. The process model is based on rigid body kinematics, while the observation model integrates the measurement equations from all sensors. After filtering, the posture monitoring and control system of the intelligent learning desk and chair outputs a smooth 5-dimensional state vector with significantly reduced noise. The posture feature extraction submodule extracts three key feature parameters from the 5-dimensional state vector. The first parameter is the pressure distribution asymmetry, calculated as follows: The posture monitoring and control system of the intelligent learning desk and chair first divides the pressure distribution matrix into a left and right half along the Y-axis, and calculates the total pressure in each half. The pressure distribution asymmetry is defined as the ratio of the absolute value of the difference between the total pressure in the left half and the total pressure in the right half to the total pressure, expressed as a percentage. The second parameter is the trunk forward tilt angle, derived from the filtered trunk pitch angle through coordinate transformation. Specifically, the trunk forward tilt angle equals 90 degrees minus the trunk pitch angle. The third parameter is the chest-abdomen distance difference, obtained directly from the chest-to-desk distance minus the abdomen-to-desk distance, in centimeters. The health benchmark comparison submodule has a set of standard healthy sitting posture range values ​​preset, determined by statistical analysis of over 1000 standard sitting posture samples. The specific range values ​​are: pressure distribution asymmetry less than or equal to 5%, trunk forward tilt angle between 85 and 95 degrees, and chest-abdomen distance difference between 3 and 8 centimeters. The posture analysis module compares the three extracted feature parameters in real time with the above preset ranges one by one. When any feature parameter exceeds its corresponding standard range, the posture monitoring and control system of the intelligent study desk and chair determines that the current posture is poor.At this point, the posture analysis module generates a posture deviation signal, which contains three fields: deviation parameter identifier, deviation direction, and deviation amount. The deviation parameter identifier indicates which feature parameter is out of range, the deviation direction indicates whether the parameter is greater than the upper limit or less than the lower limit, and the deviation amount quantifies the degree to which it exceeds the range.

[0026] The control strategy generation module generates optimal corrective control commands based on the posture deviation signal output by the posture state analysis module, combined with historical data and user-personalized parameters. The control strategy generation module comprises two main sub-modules: a strategy mapping sub-module and a personalized correction sub-module. The strategy mapping sub-module stores a preset control strategy mapping table, which defines the correspondence between posture deviation signals and initial control commands in key-value pairs. The specific mapping rules are as follows: when the deviation parameter identifier in the posture deviation signal is pressure distribution asymmetry, the initial control command is to activate the seat leveling mechanism. The command includes a baseline adjustment amount, the sign of which is determined by the deviation direction, and its magnitude is proportional to the deviation amount. When the deviation parameter identifier is torso forward tilt angle, the initial control command is to simultaneously activate the backrest tilt angle adjustment mechanism and the desktop height adjustment mechanism. The baseline adjustment amounts for both mechanisms are calculated separately based on the deviation amount and direction. When the deviation parameter identifier is chest-abdomen distance difference, the initial control command is to activate the seat fore-aft position adjustment mechanism, and its baseline adjustment amount is also calculated based on the deviation information. The personalized correction sub-module is responsible for fine-tuning the initial control commands. The personalized correction submodule accesses the user's personal profile stored in the local non-volatile memory of the posture monitoring and control system of the smart learning desk and chair. The profile contains two important types of data: historical posture data and personal physiological parameters. Historical posture data records every identified poor posture event over the past 7 days, including the event type, timestamp, deviation amount, and the effect of subsequent control commands. Personal physiological parameters include the user's pre-entered height and weight data. The personalized correction submodule uses a linear regression model to calculate the personalized adjustment coefficient. The model's input features include the 7-day average deviation of the current poor posture type, the user's height, and the user's weight. The model outputs an adjustment coefficient between 0.5 and 2.0. Multiplying this coefficient by the baseline adjustment amount in the initial control command yields the final personalized control command. The command includes detailed information such as the target actuator, target position or angle, and movement speed.

[0027] The actuator drive module receives personalized control commands from the control strategy generation module and transforms these commands into specific mechanical actions. The actuator drive module includes an instruction parsing submodule and a motor drive submodule. The instruction parsing submodule first decodes the personalized control commands, extracting the target mechanism identifier, target position coordinates, and motion speed parameters. The target mechanism identifier is used to select the specific actuator requiring the action. The target position coordinates are floating-point numbers representing the absolute position or angle the mechanism needs to reach. The motion speed parameters define the rated speed at which the motor performs the action. The motor drive submodule then generates corresponding control signals to drive the motor based on the parsed instructions. The posture monitoring and control system of the intelligent learning desk and chair controls four independent actuators, each driven by a DC servo motor and equipped with a high-precision rotary encoder to form a position closed loop. The seat leveling mechanism controls the left-right tilt angle of the seat surface through a motor. The motor is connected to the seat chassis through a worm gear reduction mechanism, and the encoder is directly mounted on the motor output shaft, providing real-time feedback on the motor rotation angle. The posture monitoring and control system of the intelligent study desk and chair converts the target tilt angle into the number of pulses required for the motor to rotate, with an adjustment range of -3 degrees to +3 degrees and a resolution of 0.1 degrees. The backrest tilt adjustment mechanism controls the backrest's pitch angle via another motor. This motor is connected to the backrest frame via a linkage mechanism, and an encoder is also used to provide feedback on the actual angle. The adjustment range is 90 degrees to 120 degrees, with a resolution of 0.5 degrees. The desktop height adjustment mechanism controls the vertical height of the desktop via a third motor. This motor drives a scissor-type lifting mechanism, and the encoder reading is converted into a height value via a lead screw. The adjustment range is 65 cm to 80 cm, with a resolution of 1 mm. The seat fore-and-aft position adjustment mechanism controls the seat's fore-and-aft displacement in the horizontal direction via a fourth motor. This motor drives a rack and pinion mechanism, and the number of encoder pulses corresponds to the linear displacement of the seat. The adjustment range is -10 cm to +10 cm, with a resolution of 1 mm. The motor drive submodule generates a pulse width modulation signal for each motor. The duty cycle of the pulse width modulation signal is proportional to the target speed of the motor, while the direction signal controls the rotation direction of the motor. The posture monitoring and control system of the intelligent learning desk and chair reads the feedback value of the encoder in real time and compares it with the target position calculated by the instruction parsing submodule, forming a closed-loop control system. This ensures that the actuator can quickly and accurately reach the target position set by the instruction, with the error controlled within 1 times the resolution of the mechanism.

[0028] refer to Figure 2The posture monitoring and control system of the intelligent learning desk and chair also includes a posture trend prediction unit, which serves as a functional extension of the posture state analysis module. The posture trend prediction unit continuously maintains a 30-bit first-in-first-out queue to store historical 5-dimensional state vectors from the most recent 30 sampling periods. The posture trend prediction unit uses a first-order autoregressive model to predict the posture state within the next 5 seconds. Specifically, for each dimension of the 5-dimensional state vector, the posture trend prediction unit fits a first-order autoregressive model based on its past 30 values. The model is formed by adding a constant term to the value of the state vector at time t minus 1, multiplied by an autoregressive coefficient, and then adding a random error term. The model parameters are estimated from historical data using the least squares method. Using the fitted model, the posture trend prediction unit recursively predicts the state vector values ​​for the next 50 time steps, i.e., 5 seconds later. Subsequently, the posture feature extraction submodule calculates future feature parameter values ​​based on these predicted state vectors. The health benchmark comparison submodule then compares these predicted feature parameters with the health benchmark range. If the prediction results indicate that a certain characteristic parameter will exceed the healthy range within the next 5 seconds, the posture analysis module will immediately output a posture deviation warning signal. The structure of the warning signal is similar to the posture deviation signal, but it includes an additional warning flag. Upon receiving the warning signal, the control strategy generation module will generate corresponding preventative control instructions. These instructions typically have smaller adjustment ranges than corrective instructions, aiming to prevent poor posture through minor, pre-adaptive adjustments, thereby upgrading the function from passive correction to active prevention.

[0029] In addition, the control strategy generation module integrates a learning optimization unit for continuously optimizing the system's control strategy. This unit runs in the background, continuously monitoring and recording the user's posture response data after each control command execution. Specifically, for each triggered control command, the learning optimization unit records the timestamp of the command execution and the command content, and continuously monitors the posture status output by the posture analysis module for the next 5 minutes. The learning optimization unit calculates the proportion of time the user maintains a standard healthy posture within this 5-minute time window, using this proportion as a quantitative indicator of the control command's effectiveness. This indicator is stored and associated with the corresponding control command. The learning optimization unit has a built-in trigger mechanism: when the smart learning desk and chair's posture monitoring and control system detects the same user triggering the same type of poor posture for the third time in the same usage scenario, the learning optimization unit is activated. After activation, the learning optimization unit retrieves the control commands and their execution effect data corresponding to the previous two instances of the same type of poor posture. The learning optimization unit analyzes these two sets of data to find the relationship between the adjustment amount of the control command and the subsequent posture improvement effect. The learning optimization unit uses gradient descent to fine-tune the parameters of the linear regression model used in the personalized correction submodule. The objective function of gradient descent is to minimize the mean squared error between the predicted posture improvement and the actual observed improvement. By iteratively adjusting the model parameters, the posture monitoring and control system of the intelligent learning desk and chair can adaptively optimize the control strategy for that specific user. This allows for the generation of more effective and accurate correction commands when faced with the same type of poor posture, thereby gradually improving the overall correction efficiency and user comfort of the posture monitoring and control system of the intelligent learning desk and chair.

[0030] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A posture monitoring and control system for an intelligent study desk and chair, characterized in that, Includes the following modules: The sitting posture data acquisition module is used to acquire multi-dimensional raw data of the user's body posture in real time; The sitting posture analysis module is used to perform fusion calculations and feature extraction based on the multi-dimensional raw data collected by the sitting posture data acquisition module, to identify the current sitting posture and determine whether it deviates from the health benchmark. The control strategy generation module is used to generate the optimal correction control instructions based on the type and degree of posture deviation, combined with historical posture data and personalized parameters. The actuator drive module is used to precisely drive the corresponding mechanical actuators of the table and chair to make adjustments according to control commands.

2. The posture monitoring and control system for the intelligent study desk and chair according to claim 1, characterized in that: The sitting posture analysis module includes a data fusion calculation submodule, a sitting posture feature extraction submodule, and a health benchmark comparison submodule; The data fusion calculation submodule uses the extended Kalman filter algorithm to perform timestamp alignment and spatial coordinate system unification on the asynchronous heterogeneous data from the pressure distribution sensing unit, the three-dimensional attitude perception unit, and the distance detection unit, and eliminates the inherent random noise and temperature drift error of each sensor, outputting a 5-dimensional state vector containing the pressure center coordinates, torso pitch angle, torso tilt angle, chest table distance, and abdominal table distance. The sitting posture feature extraction submodule extracts three key feature parameters from the 5-dimensional state vector. The first parameter is the pressure distribution asymmetry, the second parameter is the trunk forward tilt angle, and the third parameter is the chest-abdomen distance difference. The health benchmark comparison submodule has a set of standard healthy sitting posture range values ​​preset within it. These range values ​​are obtained by statistically analyzing a large number of standard sitting posture samples.

3. The posture monitoring and control system for the intelligent study desk and chair according to claim 2, characterized in that: The health benchmark comparison submodule compares the three feature parameters extracted in real time with the preset standard healthy sitting posture range values ​​one by one. When any feature parameter exceeds its corresponding standard range, the current sitting posture is determined to be a bad sitting posture, and a sitting posture deviation signal containing the specific deviation parameter identifier and the deviation amount is output.

4. The posture monitoring and control system for the intelligent study desk and chair according to claim 1, characterized in that: The control strategy generation module includes a strategy mapping submodule and a personalized correction submodule; The strategy mapping submodule stores a preset control strategy mapping table, which defines the correspondence between different types of sitting posture deviation signals and preliminary control commands. The personalized correction submodule dynamically corrects the adjustment range of the initial control command based on the user's historical sitting posture data and personal physiological parameters stored in the local non-volatile memory.

5. The posture monitoring and control system for the intelligent study desk and chair according to claim 4, characterized in that, The specific mapping rules of the strategy mapping submodule are as follows: when the sitting posture deviation signal indicates that the pressure distribution asymmetry exceeds the standard, the initial control command is to activate the seat level adjustment mechanism; when the signal indicates that the torso forward tilt angle exceeds the standard, the initial control command is to activate the backrest tilt angle adjustment mechanism and the desktop height adjustment mechanism; when the signal indicates that the chest-abdomen distance difference exceeds the standard, the initial control command is to activate the seat fore-aft position adjustment mechanism.

6. The posture monitoring and control system for the intelligent study desk and chair according to claim 4, characterized in that: The personalized correction submodule calculates the average deviation of the user's same type of poor sitting posture in the past 7 days, queries the user's pre-entered height and weight data, calculates the personalized adjustment coefficient for the user through a linear regression model, and multiplies the coefficient by the benchmark adjustment amount in the initial control command to generate the final personalized control command.

7. The posture monitoring and control system for the intelligent study desk and chair according to claim 1, characterized in that: The actuator drive module includes an instruction parsing submodule and a motor drive submodule. The instruction parsing submodule receives personalized control instructions from the control strategy generation module and decodes the personalized control instructions into specific mechanism action commands, including target mechanism identifier, target position coordinates, and motion speed parameters. The motor drive submodule generates corresponding pulse width modulation signals based on the mechanism action commands to drive the DC servo motor of the corresponding mechanism to move.

8. The posture monitoring and control system for the intelligent study desk and chair according to claim 5, characterized in that, The posture monitoring and control system of the intelligent study desk and chair controls a total of 4 actuators: the seat level adjustment mechanism controls the tilt angle of the seat surface in the left and right directions through one motor; the backrest tilt angle adjustment mechanism controls the tilt angle of the backrest through one motor; the desktop height adjustment mechanism controls the vertical height of the desktop through one motor; and the seat front and rear position adjustment mechanism controls the front and rear displacement of the seat in the horizontal direction through one motor.

9. The posture monitoring and control system for the intelligent study desk and chair according to claim 2, characterized in that: The sitting posture analysis module also includes a sitting posture trend prediction unit. The sitting posture trend prediction unit uses a first-order autoregressive model to predict changes in sitting posture within the next 5 seconds based on a historical 5-dimensional state vector sequence over a preset time period. If the prediction result shows that a certain feature parameter will exceed the health benchmark range within the next 5 seconds, the sitting posture analysis module will output a sitting posture deviation warning signal in advance.

10. The posture monitoring and control system for the intelligent study desk and chair according to claim 1, characterized in that, The control strategy generation module also integrates a learning optimization unit; the learning optimization unit continuously records the user's posture response data after each control command is executed, specifically recording the proportion of time the user returns to a standard sitting posture within 5 minutes after execution; when the same user triggers the same type of poor sitting posture for the third time in the same scenario, the learning optimization unit will analyze the execution effect data of the previous two control commands and use the gradient descent method to fine-tune the parameters of the linear regression model in the personalized correction submodule.

Images