Treadmill heart rate adaptive control method and system

Abstract

The application discloses a treadmill heart rate self-adaptive control method and system, comprising the following steps: S1: setting a target heart rate value; S2: acquiring a treadmill speed instruction and an actual user heart rate in real time according to a preset period; S3: based on the acquired historical data sequence, a discrete time model parameter for describing a current user's "speed-heart rate" dynamic relationship is identified and updated on line through a recursive least square method with a forgetting factor; S4: according to the discrete time model parameter, adaptive controller parameters are calculated based on the internal model control principle; S5: a new treadmill speed instruction is determined according to the deviation between the target heart rate value and the actual user heart rate and the adaptive controller parameters; and S6: the target speed is adjusted by using the new treadmill speed instruction, and the step S2 is returned to be circularly performed. The effect is that high-precision and high-stability full-automatic heart rate control is realized, a truly personalized adaptation is realized, and an intelligent adaptive capacity for responding to dynamic changes of user states is possessed.

Description

Technical Field

[0001] This invention belongs to the field of automatic control technology, and in particular relates to a method and system for adaptive heart rate control of a treadmill. Background Technology

[0002] Treadmills, as a widely used aerobic training device, have garnered significant attention for their heart rate-targeted training capabilities, which effectively improve cardiopulmonary function and fat-burning efficiency. A precise mathematical model of the system is the core foundation for a deep understanding of the body's physiological response to exercise load, designing high-performance controllers, ensuring training safety, and achieving personalized training.

[0003] However, existing treadmill heart rate control technology mainly relies on manual feedback adjustment modes or preset program control modes.

[0004] In manual mode, the treadmill displays the user's heart rate in real time; the user observes the heart rate reading and manually adjusts the speed or incline using the control panel to maintain the heart rate within the target range. Its drawback is:

[0005] (1) Reliance on experience and subjectivity: The effect of regulation depends heavily on the user's cognition and operational response to their own physical abilities, and ordinary users find it difficult to control it accurately.

[0006] (2) Adjustment lag and large fluctuations: There is a dual physiological and operational delay between heart rate changes and speed adjustment. Users usually start adjusting only after their heart rate has deviated significantly, causing the heart rate to fluctuate continuously throughout the training process and fail to stabilize within the target range, which seriously affects the training effect.

[0007] (3) The process is cumbersome and the experience is fragmented: Users need to constantly switch between running, observing, thinking and manual operation, which interrupts the rhythm of exercise and results in a poor experience.

[0008] For preset program control modes, treadmills have several built-in preset training programs (such as heart rate zone training and interval training). These programs automatically control the treadmill according to a pre-set, fixed speed / incline time sequence. The drawback is that:

[0009] (1) Lack of personalized adaptation: The preset program is based on the "standard population" model and does not take into account the individual differences of users. It may be insufficient for some users, while it may be too strong or even risky for others.

[0010] (2) Open-loop control, unable to cope with disturbances: The program executes according to fixed instructions, which is an "open-loop" system. It cannot sense and respond to changes in heart rate caused by factors such as fatigue, dehydration, and distraction, resulting in an increasing deviation between the actual heart rate and the target value.

[0011] (3) It is essentially speed control rather than heart rate control: its control goal is to complete the preset speed curve, rather than to stabilize the heart rate at a certain target value.

[0012] In summary, the core problem with existing methods is that "heart rate," a physiological output that needs precise control, is not truly used as a closed-loop feedback signal for the control system. They either rely on unreliable manual closed loops or lack any closed loop at all, resulting in low control accuracy, poor personalization, and a subpar user experience. Summary of the Invention

[0013] In view of this, the present invention first provides a treadmill heart rate adaptive control method, which mainly solves the following three core problems existing in the prior art:

[0014] (1) Solve the problems of low control accuracy and poor stability.

[0015] To address the dual lag (physiological delay + operational delay) and subjective arbitrariness inherent in the existing "manual observation-manual adjustment" model, this invention aims to solve the problem of how to achieve fully automatic, high-precision, and highly stable heart rate control. Specifically, it addresses how to replace manual intervention with technological means, eliminating human observation, decision-making, and operational steps, enabling the heart rate to be automatically, quickly, and smoothly regulated and stabilized within a preset target value or range, avoiding large fluctuations and oscillations, and ensuring the accuracy of the training load.

[0016] (2) Solving the problem of lack of personalized adaptation of control strategies

[0017] Addressing the issue of the fixed speed curve applied in the existing "preset program-open loop execution" model, this invention aims to solve the problem of providing customized control for each specific user. Specifically, it addresses how to enable the control system to automatically identify and learn the unique, dynamic "speed-heart rate" physiological response model parameters of the current user online. This allows the generation of control commands (treadmill speed) to be truly based on the user's individual real-time physiological characteristics, achieving a leap from "one speed for everyone" to "one strategy for each person."

[0018] (3) Solving the problem that the control system cannot adapt to dynamic changes.

[0019] To address the issue of continuously changing physiological states and response characteristics of users during the same training session due to factors such as fatigue, warm-up, and dehydration, this invention aims to solve the problem of how to enable the control system to possess online self-optimization capabilities. Specifically, based on the already achieved personalized model identification, how to further design a control algorithm that can automatically update its internal control parameters (such as the internal model inverse model and filter coefficients) according to model changes or real-time control effects, actively controlling the speed of the treadmill, thereby continuously maintaining optimal control performance throughout the entire exercise process and coping with the dynamic drift of the user's physiological characteristics.

[0020] To achieve the above objectives, the specific technical solution adopted by the present invention is as follows:

[0021] A treadmill heart rate adaptive control method, the key of which includes the following steps:

[0022] S1: Set the target heart rate value;

[0023] S2: Real-time acquisition of treadmill speed commands and user's actual heart rate according to preset cycles;

[0024] S3: Based on the acquired historical data sequence, the discrete-time model parameters used to describe the current user's "speed-heart rate" dynamic relationship are identified and updated online using the recursive least squares method with a forgetting factor.

[0025] S4: Calculate the adaptive controller parameters based on the internal model control principle according to the discrete-time model parameters;

[0026] S5: Determine a new treadmill speed command based on the deviation between the target heart rate value and the user's actual heart rate, as well as the adaptive controller parameters obtained in step S4.

[0027] S6: Adjust the target speed using the new treadmill speed command, and return to step S2 to repeat the process.

[0028] Optionally, the discrete-time model used to describe the current user's "speed-heart rate" dynamic relationship is a controlled autoregressive model, whose expression is: ;

[0029] in, for Heart rate at all times for The treadmill speed command at any given moment. For transfer functions, and Let be the polynomial to be identified, and define it as:

[0030]

[0031]

[0032] Model parameters ;

[0033] This indicates a shift operator that satisfies the relation ;

[0034] and The order of the polynomial is selected based on the system characteristics; and These are the model parameters to be identified.

[0035] Optionally, in the recursive least squares method with a forgetting factor, the recursive form is:

[0036] Gain vector calculation follows:

[0037]

[0038] Parameter estimation updates are based on:

[0039]

[0040] The covariance matrix is ​​updated according to:

[0041]

[0042] in: For historical data sequences, for Heart rate at all times Forgetting factor, The identity matrix and the initial parameter vector The initial covariance matrix is ​​configured as a zero vector or based on empirical values. , The initial value configured.

[0043] Optionally, the forgetting factor Based on the model's posterior residuals Dynamic adjustment, specifically: ,in, The lower limit of the forgetting factor is the posterior residual of the model. .

[0044] Optionally, by constructing a historical data sequence using second-order historical data, the obtained discrete-time model can be updated to a second-order model through step S3.

[0045] Optionally, the adaptive controller is an internal model controller, whose transfer function is... The inverse of the invertible part of the discrete-time model With low-pass filter The low-pass filter is configured in series, and the adjustment coefficient of the low-pass filter is... The treadmill speed command is dynamically adjusted based on the residuals identified by the model, and the final speed command is as follows:

[0046] Confirmed, including feedback signals Defined as actual heart rate output Compared with the internal model prediction output The difference between them.

[0047] Optionally, the low-pass filter is of first-order form; if the model identification residual is small, the size can be reduced. To improve response speed; if the residual is large, increase Enhance system robustness and prevent drastic speed fluctuations.

[0048] Based on the above method, the present invention also provides a treadmill heart rate adaptive control system, which is equipped with a speed-adjustable motor, a speed sensor, a control panel and a central controller unit. The key features are: the user sets a target heart rate value through the control panel, the central controller unit obtains the speed feedback of the treadmill through the speed sensor, the central controller unit obtains the user's actual heart rate through the heart rate detection module, and the central controller unit also performs control according to the treadmill heart rate adaptive control method described above.

[0049] Optionally, the user configures a heart rate detection device and connects it to the heart rate detection module via wired or wireless means. The control panel is used to display system status including real-time speed and real-time heart rate, and is also used to store model parameters, motor speed and heart rate data.

[0050] Furthermore, this invention also protects a treadmill, the key feature of which is that it is equipped with the aforementioned treadmill heart rate adaptive control system.

[0051] The significant effects of this invention are:

[0052] (1) Achieved high-precision and high-stability fully automatic heart rate control

[0053] By directly using the heart rate signal as the real-time feedback input to the controller and automatically calculating and outputting speed commands, the system completely eliminates the double delay and subjective error caused by the "manual observation-decision-manual operation" chain. The system can fine-tune the treadmill speed with a millisecond-level response speed. As the human body follows the changes in the treadmill speed, the heart rate can be quickly and smoothly adjusted and stabilized near the target value, with significantly smaller fluctuations than manual control, achieving truly precise load control.

[0054] (2) It has achieved true personalized adaptation.

[0055] The core of this invention lies in its ability to automatically and online identify the current user's "speed-heart rate" dynamic model. This means the control system no longer relies on preset, fixed programs based on average groups, but instead constructs a personalized physiological response model for each user and generates control commands based on this model. It automatically provides higher speeds to physically strong individuals and appropriate speed adjustments to those with weaker constitutions, ensuring that different treadmill speeds can be adjusted to the same target heart rate for different users, achieving a fundamental shift from "device controller" to "device adaptor."

[0056] (3) It has the intelligent adaptability to deal with dynamic changes in user status.

[0057] Traditional controllers have fixed parameters and cannot adapt to the different characteristics of users from warm-up, steady state to fatigue stage. The adaptive controller of this invention can adjust the internal model and filter bandwidth of the controller in real time according to the online identified heart rate regulation model parameters of the exerciser, ensuring that the system can maintain the optimal treadmill speed throughout the entire exercise, regardless of the user's physiological state. This is something that preset programs or fixed parameter controllers cannot achieve.

[0058] (4) It significantly reduces the cognitive load on users and improves the sports experience and safety.

[0059] Simplified interaction, focusing on the exercise itself: Users no longer need to be distracted by watching the gauges or thinking about how to adjust the speed; they can simply passively follow the treadmill's automatically adjusted speed. This greatly reduces the barrier to entry and cognitive burden, making the exercise process more coherent and focused, which is especially beneficial for rehabilitation patients or beginners to perform precise training while ensuring safety.

[0060] Built-in safety monitoring: The system continuously monitors the heart rate response. If an abnormal spike occurs or the heart rate cannot be adjusted (which may indicate discomfort), this abnormal mode can serve as an early warning signal, providing an intelligent basis for integrating emergency deceleration or shutdown functions.

[0061] (5) It provides a quantifiable technical basis for optimizing training results.

[0062] Ensure training specificity: By keeping the heart rate stable within a specific range (fat-burning range, aerobic endurance range) for an extended period, it is possible to strictly ensure that each training session accurately achieves the preset physiological goals (maximizing fat consumption or improving cardiorespiratory threshold), avoiding the reduction in training effectiveness due to heart rate fluctuations.

[0063] Generate personalized data assets: The user's personal model and control process data identified by the system are valuable digital assets reflecting their cardiopulmonary function and fatigue status. They can be used to track changes in physical fitness over a long period of time and adjust training plans in a personalized manner, realizing the upgrade from "experience-based training" to "data-driven training". Attached Figure Description

[0064] Figure 1 This is a diagram of the treadmill heart rate adaptive control system architecture provided in an embodiment of the present invention;

[0065] Figure 2 This is a flowchart of the real-time heart rate motion model identification and control provided in an embodiment of the present invention. Detailed Implementation

[0066] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. The embodiments are given for illustrative purposes only and should not be construed as limiting the present invention. The accompanying drawings are for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0067] This invention provides a treadmill heart rate adaptive control method and system, the core architecture of which is a real-time heart rate model parameter identification and adaptive control system. The heart rate control loop is composed of an online model identification module and an adaptive controller module connected in series, forming an integrated intelligent closed loop of "identification-control".

[0068] like Figure 1 The treadmill real-time heart rate control system shown primarily comprises the following hardware components: a treadmill (including an adjustable motor and speed sensor), a heart rate monitoring device (containing a heart rate belt or optical sensor, connected via wired or wireless means), a central control unit, a motor driver, and a control panel. The control panel is connected to the central control unit via wires and is used to set the target heart rate and display system status, including real-time speed and real-time heart rate. The control panel can also store model parameters, motor speed, and heart rate data. The heart rate monitoring device collects heart rate data and processes noise; the processed data is then transmitted to the central control unit. After collecting the treadmill speed and heart rate, the central control unit performs model identification and adaptive control law calculation.

[0069] The block diagram of the real-time heart rate motion model identification controller is as follows: Figure 2 As shown, the real-time identification and control system mainly consists of a treadmill motion and heart rate model. The system consists of model parameter identification, an internal model controller designer, and a controller. First, the target heart rate is set via the control panel. After startup, the central control unit outputs a pseudo-random signal (PRBS) to control the motor's speed adjustment for the first 300 seconds, while simultaneously collecting heart rate data to identify the initial exercise heart rate model. Once the central control unit obtains the initial model parameters, it enters the adaptive control phase: based on the set heart rate, the controller designer updates the control parameters in real time according to the internal model principle, controlling the motor driver to adjust the treadmill speed. The athlete follows the treadmill's speed; because the exercise heart rate model is known and updated in real time, the treadmill allows the athlete's heart rate to quickly reach the set value after speed adjustment.

[0070] The above method can be embedded into the central control unit through software programming. The basic control flow includes:

[0071] S1: Set target heart rate value ;

[0072] S2: Obtain treadmill speed commands in real time according to a preset cycle (1 second). and the user's actual heart rate ;

[0073] S3: Based on the acquired historical data sequence The discrete-time model parameters describing the current user's "speed-heart rate" dynamic relationship are identified and updated online using recursive least squares with a forgetting factor (FFRLS). ;

[0074] S4: Calculate the adaptive controller parameters based on the internal model control principle according to the discrete-time model parameters;

[0075] S5: Based on the deviation between the target heart rate value and the user's actual heart rate, and the adaptive controller parameters obtained in step S4. Determine the new treadmill speed command ;

[0076] S6: Adjust the target speed using the new treadmill speed command, and return to step S2 to repeat the process.

[0077] This invention uses real-time data acquisition to inversely deduce physiological parameters that characterize the current user (such as the cardiorespiratory endurance time constant). Physiological gain The system state parameters are used to dynamically adjust the controller for optimal heart rate tracking. In practice, since the treadmill-human coupling system consists of first-order motor dynamics and first-order physiological dynamics connected in series, the overall system exhibits second-order linear system characteristics. Based on this, the polynomial order of the system can be pre-set. This directly corresponds to the second-order differential equation: The discretized form; the sampling period can be configured from 1 second to 2 seconds to match the physiological time constant of human heart rate changes; the target heart rate is the user-defined expected exercise heart rate, and the initial parameter vector of the recursive least squares (FFRLS) method is set. Zero vector or empirical value; initial covariance matrix ( Take the larger number, such as ), to represent the uncertainty of the parameters at the initial time; set the initial value of the forgetting factor. and lower limit In addition, the initial adjustment coefficient of the internal model control filter is set. .

[0078] When the new sampling time Upon arrival, the central control unit performs the following operations:

[0079] Input collection : Reads the current speed command of the treadmill motor. Physically, this represents the external work load applied to the human body.

[0080] Acquisition and Output The system obtains the user's actual heart rate through a heart rate sensor. Physically, this represents the metabolic response of the human cardiovascular system to load.

[0081] This constitutes the input / output data pair used for identification. .

[0082] For step S3, the aim is to "reverse learn" the user's current physiological state using mathematical methods. Discrete parameters This implicitly includes the user's weight, exercise efficiency, and endurance level, and calls the parameter estimator module:

[0083] (a) Constructing a regression vector: Constructing an information vector using historical data:

[0084]

[0085] Here we utilize and The data at each moment is precisely for capturing the second-order inertial characteristics of the system.

[0086] (b) Update the gain vector:

[0087] Calculate the correction gain:

[0088]

[0089] (c) Update parameter estimates: Correct model parameters based on prediction errors:

[0090]

[0091] Obtained here This refers to the user's "speed-heart rate" transfer function parameters extracted from the data at the current moment.

[0092] (d) Update the covariance matrix:

[0093]

[0094] (e) Update the forgetting factor:

[0095] Based on the dynamic calculation of posterior residuals .

[0096] When a user's physiological state undergoes a sudden change (such as entering a period of fatigue), it affects the model parameters. When there are drastic changes, the residuals increase, and the algorithm automatically reduces... This will accelerate the tracking of new physiological characteristics.

[0097] (f) Extracting the model polynomial:

[0098] vector Restored to ARX model polynomial and Thus, the second-order model can be identified:

[0099]

[0100] For step S4, based on the identified physiological model, a controller is designed using the internal model control (IMC) principle to overcome the hysteresis between the motor and human physiology. Specific steps include:

[0101] (a) Constructing the inverse model: Calculating the current identification model The inverse model. Physical significance: The role of the inverse model is to mathematically cancel out the inertia of the motor and physiological inertia, thereby theoretically achieving an instantaneous response of heart rate.

[0102] (b) Design a low-pass filter: Construct the filter parameter The physical meaning of this is that it determines the closed-loop bandwidth of the control system. If the model identification residual is small, it indicates that the model is accurate and can reduce [the impact of errors]. To improve response speed; if the residual is large, it indicates the presence of nonlinear interference, and the response speed should be increased. It can enhance system robustness and prevent drastic speed fluctuations.

[0103] (c) Composite Controller: Calculate the final controller transfer function: .

[0104] Step S5 specifically includes:

[0105] (a) Estimating generalized perturbations: Computing model prediction output And seek the differences .

[0106] Physical meaning: This reflects aspects that the model failed to explain, including external environmental disturbances, heart rate fluctuations caused by psychological factors, and model mismatch errors.

[0107] (b) Calculate the control law:

[0108] This step is controlled by the controller. The corrected target (target heart rate minus the perturbation estimate) is calculated to determine the treadmill speed that should be applied at the current moment.

[0109] (c) Execute speed adjustment: Calculate the digital quantity It is converted into a motor drive signal (such as PWM duty cycle) to change the belt speed.

[0110] Finally, wait for the next sampling cycle, and through repeated cyclic control, achieve adaptive optimal control throughout the entire process.

[0111] In summary, this embodiment proposes a treadmill that uses the system described above and is controlled according to the high-precision, high-stability fully automatic heart rate adaptive control method proposed in this invention. This achieves true personalized adaptation, significantly reduces the cognitive load on users, and improves the exercise experience and safety.

[0112] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A treadmill heart rate adaptive control method, characterized in that, Includes the following steps: S1: Set the target heart rate value; S2: Real-time acquisition of treadmill speed commands and user's actual heart rate according to preset cycles; S3: Based on the acquired historical data sequence, the discrete-time model parameters used to describe the current user's "speed-heart rate" dynamic relationship are identified and updated online using the recursive least squares method with a forgetting factor. S4: Calculate the adaptive controller parameters based on the internal model control principle according to the discrete-time model parameters; S5: Determine a new treadmill speed command based on the deviation between the target heart rate value and the user's actual heart rate, as well as the adaptive controller parameters obtained in step S4. S6: Adjust the target speed using the new treadmill speed command, and return to step S2 to repeat the process; The discrete-time model used to describe the dynamic relationship between the current user's "speed and heart rate" is a controlled autoregressive model, and its expression is: ; in, for Heart rate at all times for The treadmill speed command at any given moment. For transfer functions, and Let be the polynomial to be identified, and define it as: Model parameters ; This indicates a shift operator that satisfies the relation ; and The order of the polynomial is selected based on the system characteristics; and These are the model parameters to be identified; The adaptive controller is an internal model controller, and its transfer function is... The inverse of the invertible part of the discrete-time model With low-pass filter The low-pass filter is configured in series, and the adjustment coefficient of the low-pass filter is... The treadmill speed command is dynamically adjusted based on the residuals identified by the model, and the final speed command is as follows: Confirmed, including feedback signals Defined as actual heart rate output Compared with the internal model prediction output The difference between them.

2. The treadmill heart rate adaptive control method according to claim 1, characterized in that: In the recursive least squares method with a forgetting factor, the recursive form is: Gain vector calculation follows: Parameter estimation updates are based on: The covariance matrix is ​​updated according to: in: For historical data sequences, for Heart rate at all times Forgetting factor, The identity matrix and the initial parameter vector The initial covariance matrix is ​​configured as a zero vector or based on empirical values. , The initial value configured.

3. The treadmill heart rate adaptive control method according to claim 2, characterized in that: The forgetting factor Based on the model's posterior residuals Dynamic adjustment, specifically: ,in, The lower limit of the forgetting factor is the posterior residual of the model. .

4. The treadmill heart rate adaptive control method according to claim 3, characterized in that: By constructing a historical data sequence using second-order historical data, the resulting discrete-time model is updated to a second-order model through step S3.

5. The treadmill heart rate adaptive control method according to claim 1, characterized in that: The low-pass filter is of first-order form. If the model identification residual is small, the value is reduced. To improve response speed; if the residual is large, increase Enhance system robustness and prevent drastic speed fluctuations.

6. A treadmill heart rate adaptive control system, comprising a speed-regulating motor, a speed sensor, a control panel, and a central controller unit on the treadmill, characterized in that: The user sets a target heart rate value through the control panel, the central controller unit obtains the speed feedback of the treadmill through the speed sensor, the central controller unit obtains the user's actual heart rate through the heart rate detection module, and the central controller unit also performs control according to the treadmill heart rate adaptive control method according to any one of claims 1-5.

7. The treadmill heart rate adaptive control system according to claim 6, characterized in that: The user configures the heart rate detection device and connects it to the heart rate detection module via wired or wireless means. The control panel is used to display the system status, including real-time speed and real-time heart rate, and also to store model parameters, motor speed, and heart rate data.

8. A treadmill, characterized in that: The treadmill is equipped with the heart rate adaptive control system as described in claim 6 or 7.

Images