A control method for closed-loop human-computer interaction device and closed-loop human-computer interaction device

Abstract

The present application discloses a control method of a closed-loop human-computer interaction device and a closed-loop human-computer interaction device, wherein the control method of the closed-loop human-computer interaction device comprises: when a user pulls a pull rope, a measuring component obtains the user's electromyographic information; the measuring component obtains fatigue index information according to the electromyographic information and sends it to a mechanical arm; the mechanical arm obtains force information according to the fatigue index information, and applies force to the pulley assembly according to the force information, so as to counteract the force of the user pulling the pull rope; when the mechanical arm determines that the current state of the user is a fatigue state according to the fatigue index information, the mechanical arm releases the pulley assembly. The present application can avoid direct or cumulative injuries that may be caused by improper fitness, thereby improving the fitness training effect.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent robots, and in particular to a control method for a closed-loop human-computer interaction device and a closed-loop human-computer interaction device. Background Art

[0002] Sports fatigue, a phenomenon involving the decline of muscle and nerve center functions, is particularly prominent during exercise, manifesting as distraction, weakened muscle strength and endurance, and decreased central nervous system regulation. For fitness athletes, it is crucial to maintain high concentration and stable muscle strength during weight training. Once the muscles and nervous system fall into a state of fatigue, it will be difficult to maintain efficient training, and forcing it to continue will significantly increase the risk of sports injuries.

[0003] At present, due to the lack of face-to-face professional guidance, users may encounter problems such as irregular movements and unreasonable training plans during exercise, which may affect the exercise effect and even bring safety risks.

[0004] Therefore, the prior art still needs to be improved and developed. Summary of the invention

[0005] The main purpose of the present application is to provide a control method for a closed-loop human-computer interaction device and a closed-loop human-computer interaction device, aiming to solve the problem in the prior art that improper fitness may cause injuries and lead to poor fitness training effects.

[0006] A first aspect of an embodiment of the present application provides a control method for a closed-loop human-computer interaction device, wherein the closed-loop human-computer interaction device includes a pull rope, a pulley assembly, a measuring assembly and a robotic arm, wherein the pull rope, the pulley assembly and the robotic arm are connected in sequence, and the measuring assembly is connected to the robotic arm; the control method for the closed-loop human-computer interaction device includes: when the user pulls the pull rope, the measuring assembly obtains the user's electromyographic information; the measuring assembly obtains fatigue index information based on the electromyographic information and sends it to the robotic arm; the robotic arm obtains force information based on the fatigue index information, and applies force to the pulley assembly according to the force information, so as to counteract the force of the user pulling the pull rope; if the robotic arm determines that the user's current state is a fatigue state based on the fatigue index information, the robotic arm releases the pulley assembly.

[0007] Optionally, in one embodiment of the present application, the electromyographic information includes a target electromyographic signal, a real-time electromyographic signal and frequency data; when the user pulls the pull rope, the measurement component obtains the user's electromyographic information, specifically including: when the user pulls the pull rope with a target force, the measurement component obtains the user's target electromyographic signal; when the user pulls the pull rope with a current force, the measurement component obtains the user's real-time electromyographic signal and real-time frequency data.

[0008] Optionally, in one embodiment of the present application, the fatigue index information includes time domain evaluation parameters and frequency domain evaluation parameters; the measurement component obtains fatigue index information based on the electromyographic information, specifically including: calculating the root mean square value based on the real-time electromyographic signal, and obtaining the frequency domain evaluation parameters based on the real-time frequency data; the measurement component obtains the time domain evaluation parameters based on the target electromyographic signal and the root mean square value.

[0009] Optionally, in one embodiment of the present application, the force information includes force magnitude and force direction; the robotic arm obtains the force information based on the fatigue index information, specifically: the robotic arm obtains the force magnitude and force direction of the robotic arm based on the time domain evaluation parameters and the frequency domain evaluation parameters.

[0010] Optionally, in one embodiment of the present application, the robotic arm determines that the current state of the user is a fatigue state based on the fatigue index information, specifically including: the robotic arm calculates instantaneous fatigue characteristics in adjacent preset time periods based on the fatigue index information; the robotic arm determines the instantaneous state of the user based on the instantaneous fatigue characteristics; within a preset period, if the robotic arm determines that the number of instantaneous states as instantaneous fatigue states is greater than a preset value, it determines that the current state of the user is a fatigue state.

[0011] Optionally, in one embodiment of the present application, the instantaneous fatigue characteristic includes a first difference and a second difference; the robotic arm calculates the instantaneous fatigue characteristic in adjacent preset time periods based on the fatigue index information, specifically: the robotic arm calculates the first difference of the time domain evaluation parameters in adjacent preset time periods and the second difference of the frequency domain evaluation parameters in adjacent preset time periods.

[0012] Optionally, in one embodiment of the present application, the robotic arm determines the instantaneous state of the user based on the instantaneous fatigue characteristic, specifically including: if the robotic arm determines that the first difference and the second difference are both greater than zero, then it determines that the instantaneous state of the user is an instantaneous fatigue state; if the robotic arm determines that the first difference is less than or equal to zero, then the instantaneous state of the user is an instantaneous non-fatigue state.

[0013] A second aspect of an embodiment of the present application also provides a closed-loop human-computer interaction device for implementing the control method of a closed-loop human-computer interaction device described in any one of the above-mentioned schemes, wherein the closed-loop human-computer interaction device includes a pull rope, a pulley assembly, a measuring assembly and a robotic arm, the pull rope, the pulley assembly and the robotic arm are connected in sequence, and the measuring assembly is connected to the robotic arm; when a user pulls the pull rope, the robotic arm applies force to the pulley assembly to counteract the force of the user pulling the pull rope.

[0014] Optionally, in one embodiment of the present application, the pulley assembly includes a fixed plate, a skateboard, a connecting rope, a first pulley, a second pulley, a third pulley and a fixed wheel, the skateboard is slidably connected to the fixed plate, the first pulley and the third pulley are connected to the skateboard, the second pulley and the fixed wheel are connected to the fixed plate, the fixed wheel is located on the side of the fixed plate away from the second pulley, one end of the connecting rope is fixed to the side close to the second pulley, and the connecting rope is connected to the robotic arm through the third pulley, the second pulley and the first pulley in sequence, one end of the pull rope is fixed to the skateboard, and the pull rope is connected to the fixed wheel, and the other end of the pull rope is used to be pulled by a user.

[0015] Optionally, in one embodiment of the present application, the closed-loop human-computer interaction device also includes a breathing light, the measuring component is connected to the breathing light, and the electromyographic information collected by the measuring component is sent to the breathing light so that the breathing light flashes according to the electromyographic information.

[0016] Beneficial effects: The present application provides a control method for a closed-loop human-computer interaction device and a closed-loop human-computer interaction device. The present application accurately captures and adjusts the force size and direction in real time according to the user's electromyographic signals by combining static force under the stability of the user's joints with dynamic adaptive adjustment of the robotic arm, and stops running when it determines that the user has reached a fatigue state, thereby avoiding direct or cumulative injuries that may be caused by inappropriate fitness, thereby improving the fitness training effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a structural schematic diagram of a preferred embodiment of the closed-loop human-computer interaction device of the present application;

[0019] Figure 2 It is a flow chart of a preferred embodiment of the control method of the closed-loop human-computer interaction device of the present application;

[0020] Figure 3 It is a flow chart of myoelectric control in a preferred embodiment of the control method of the closed-loop human-computer interaction device of the present application;

[0021] Figure 4It is a flow chart of muscle fatigue judgment in a preferred embodiment of the control method of the closed-loop human-computer interaction device of the present application.

[0022] Description of reference numerals:

[0023] 10. Pull rope; 20. Robotic arm; 31. Fixed plate; 32. Slide plate; 33. Connecting rope; 34. First pulley; 35. Second pulley; 36. Third pulley; 37. Fixed wheel.

[0024] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical scheme and effect of the present application clearer and more specific, the technical scheme in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. The described embodiments are only possible technical implementations of the present application, not all possible implementations. Based on the embodiments in the present application, those skilled in the art can completely combine the embodiments of the present application to obtain other embodiments without creative work, and these embodiments are also within the scope of protection of the present application.

[0026] The following describes the control method of the closed-loop human-computer interaction device and the closed-loop human-computer interaction device of the embodiment of the present application with reference to the accompanying drawings. In view of the problem in the above-mentioned related art that improper fitness may cause injuries, resulting in poor fitness training effects, the present application provides a control method for a closed-loop human-computer interaction device. In this control method, by ensuring the static force under the stability of the user's joints and combining it with the dynamic adaptive adjustment of the robotic arm, the force size and direction are accurately captured and adjusted in real time according to the user's electromyographic signals, and the operation is stopped when it is determined that the user reaches a fatigue state, thereby avoiding direct or cumulative injuries that may be caused by improper fitness, thereby improving the fitness training effect. As a result, the technical problem in the related art that improper fitness may cause injuries, resulting in poor fitness training effects, is solved.

[0027] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0028] First, the following will be combined with the attached Figure 1 , the closed-loop human-computer interaction device of an embodiment of the present application is introduced.

[0029] like Figure 1As shown, an embodiment of the present application provides a closed-loop human-computer interaction device, which includes a pull rope 10, a pulley assembly, a measuring assembly (not shown in the figure) and a robotic arm 20, wherein the pull rope 10, the pulley assembly and the robotic arm 20 are connected in sequence, and the measuring assembly is connected to the robotic arm 20; when the user pulls the pull rope 10, the robotic arm 20 applies force to the pulley assembly to counteract the force of the user pulling the pull rope 10.

[0030] Specifically, the measuring component is connected to the user's arm to measure the electromyographic signal.

[0031] In one embodiment of the present application, Figure 1 As shown, the pulley assembly includes a fixed plate 31, a skateboard 32, a connecting rope 33, a first pulley 34, a second pulley 35, a third pulley 36 and a fixed wheel 37, the skateboard 32 is slidably connected to the fixed plate 31, the first pulley 34 and the third pulley 36 are connected to the skateboard 32, the second pulley 35 and the fixed wheel 37 are connected to the fixed plate 31, the fixed wheel 37 is located on the side of the fixed plate 31 away from the second pulley 35, one end of the connecting rope 33 is fixed to the side close to the second pulley 35, and the connecting rope 33 is connected to the robot arm 20 through the third pulley 36, the second pulley 35 and the first pulley 34 in sequence, one end of the pull rope 10 is fixed on the skateboard 32, and the pull rope 10 is connected to the fixed wheel 37, and the other end of the pull rope 10 is used to be pulled by the user.

[0032] Specifically, a lifting groove is provided on the fixed plate 31, and the slide plate 32 is lifted and lowered on the fixed plate 31 through the lifting groove. A first fixing ring is provided at the upper right end of the fixed plate 31, and the connecting rope 33 is connected to the robot arm 20 after passing through the first fixing ring. A second fixing ring is provided at the lower left end of the fixed plate 31, and is pulled by the user after being stretched through the second fixing ring.

[0033] It should be noted that by calculating the electromyographic characteristic data, the characteristics that meet the corresponding force are found, and the rise or fall of the mechanical arm 20 is controlled accordingly to drive the movement of the pulley assembly. The user can stimulate the forearm and upper arm muscles by tightening the tension rope (pull rope 10) fixed on the movable pulley assembly, that is, without the need for traditional arm flexion and extension, to achieve efficient muscle training and avoid the joint pressure that may be caused by traditional dynamic flexion and extension movements.

[0034] It is understandable that the present application can achieve the adjustment of dynamic exercise load. Users can adjust the force according to their physical fitness level and training goals to achieve personalized matching of strength. At the same time, the additional pull rope (pull rope 10) can more intuitively show the change of force. Through this design, the fitness needs of people of different ages and physical fitness levels can be met.

[0035] Moreover, the pulley assembly can change the force size and direction of the mechanical arm 20, so that the strength of the mechanical arm 20 matches that of the human body, and by dynamically adjusting the exercise load, it can ensure that the exercise intensity is sufficient and effectively prevent fatigue damage caused by excessive exercise. In addition, when the mechanical arm 20 is subjected to a large force, it can be effectively buffered by the pulley structure, protecting the mechanical arm 20 from damage and extending its service life.

[0036] The present invention realizes fine adjustment of strength level through the design of adjustable pulley assembly, so that both beginners and senior fitness enthusiasts can find the exercise intensity that suits them. It fully takes into account the members of different ages and physical conditions in modern families, ensuring wide applicability.

[0037] In one embodiment of the present application, the closed-loop human-computer interaction device also includes a breathing light, the measuring component is connected to the breathing light, and the electromyographic information collected by the measuring component is sent to the breathing light so that the breathing light flashes according to the electromyographic information.

[0038] Specifically, the smart breathing light is based on the single-chip minimum system, on which a Bluetooth module and an adjustable RGB light are installed. When the user is exercising, the collected muscle strength data will be sent to the Bluetooth of the single-chip microcomputer in real time, and the Bluetooth will be sent to the single-chip microcomputer body through the serial port. The single-chip microcomputer will output different PWM waves (pulse width modulation waveforms) to the three pins of the RGB light according to the different muscle strengths, realizing flashing lights of different frequencies formed by several color combinations in RGB to achieve the visual effect of the breathing light. This application can monitor the exercise status in real time, and you can understand your exercise status in real time by observing the changes in the breathing light, which enhances the fun and interactivity of the exercise.

[0039] This application breaks through the limitations of fitness equipment, eliminates long-standing pain points such as poor exercise effects, difficult to control safety risks, narrow applicable population and insufficient interactivity, and introduces cutting-edge bioelectric signal technology. It collects human surface electromyographic signals with high precision and conducts real-time and accurate data analysis, which is used as intelligent instructions to drive the movement of the robotic arm, ensuring that each movement can accurately locate and stimulate the key muscle groups of the forearm and upper arm, achieving personalized and efficient muscle training effects, while effectively avoiding the risk of fatigue injuries caused by improper exercise.

[0040] This application forms a closed-loop control system for human-machine interaction through the real-time detection of biological signals and the linkage of mechanical devices. This system establishes a neural information measurement and processing channel and a human-machine information channel. Through the coordinated work of electromyography acquisition equipment, adjustable pulley assemblies, and joint card mechanical arms, and supplemented by intelligent breathing lights and user interfaces, it enhances user immersion and convenience, realizes the machine's deep perception of the human body state and intelligent interaction, and provides users with a personalized fitness experience.

[0041] Next, a control method for a closed-loop human-computer interaction device proposed according to an embodiment of the present application is described with reference to the accompanying drawings.

[0042] like Figure 2 As shown, the control method of the closed-loop human-computer interaction device includes the following steps:

[0043] In step S101, when the user pulls the pull rope, the measurement component obtains the user's electromyographic information.

[0044] In a possible implementation, the electromyographic information includes a target electromyographic signal, a real-time electromyographic signal, and frequency data. When the user pulls the pull rope with a target force, the measurement component obtains the user's target electromyographic signal; when the user pulls the pull rope with a current force, the measurement component obtains the user's real-time electromyographic signal and real-time frequency data.

[0045] It can be understood that the target electromyographic signal is the surface electromyographic signal collected corresponding to when the user uses the maximum force.

[0046] Specifically, the surface electromyographic signal sEMG is an AC signal of a one-dimensional time action potential sequence, and the amplitude is generally proportional to the strength of muscle movement. Ideally, sEMG is a non-stationary micro-electric signal. This application uses a 10Hz 5th-order IIR (Infinite Impulse Response) low-pass causal filter and a 50Hz notch filter to filter out interference. After completing the frequency domain analysis, the signal is analyzed and processed in the time domain, and the collected upper arm muscle group signal is converted to an analog-to-digital AD value (Analog-to-Digital Conversion) to obtain a time domain diagram of the signal. According to the time domain diagram, it can be seen that the collected signals are positive and negative, and because some average parameter correlation coefficients of the signal need to be calculated later, the positive and negative components cancel each other out. Therefore, this application further performs a signal rectification operation. After rectification, the signal has a relatively obvious average power envelope, which also shows that the signal time domain has obvious energy-related characteristics.

[0047] In step S102, the measuring component obtains fatigue index information according to the electromyographic information and sends it to the robotic arm.

[0048] In a possible implementation, the fatigue index information includes a time domain evaluation parameter and a frequency domain evaluation parameter. The measurement component calculates a root mean square value according to the real-time electromyographic signal and obtains a frequency domain evaluation parameter according to the real-time frequency data; the measurement component obtains a time domain evaluation parameter according to the target electromyographic signal and the root mean square value.

[0049] It is understandable that in the fatigue analysis of surface electromyographic signals based on characteristic parameter analysis, there are characteristic indicators such as MAV (mean absolute value), RMS (root mean square value), ZC (zero crossing rate), IEMG (integrated electromyographic value), AR (autoregression) based on the time domain analysis method, and MNF (mean frequency), MDF (median frequency), MIF (instantaneous frequency) and other characteristic indicators based on frequency domain analysis. This application adopts a joint analysis method of time domain and frequency domain to perform feature processing on electromyographic data. In the above time-frequency domain features, the features that can clearly reflect the change in the amplitude of the surface electromyographic signal are RMS and IEMG. The characteristic indicators that clearly reflect the electromyographic signal in the frequency domain are MNF and MDF. In the time domain features, it is obvious that IMEG and RMS have the same trend of change and reflect the same characteristics. This application only selects RMS as the calculation indicator. In the frequency domain features, assuming that a sufficiently short window function is selected to extract these features, then in a short time, MNF and MDF also have the same trend of change and reflect the same characteristics. Therefore, this application also only selects MNF as the calculation indicator.

[0050] The calculation formula of the root mean square value RMS is as follows:

[0051] ;

[0052] in, Indicates the length of the time window, indicating the time window from a certain starting point ( ) to the current point ( ) time interval; represents the integrating variable, which represents small increments of time; Indicates the current time or the end point of the analyzed time window, It is the surface electromyography signal.

[0053] The calculation formula of mean frequency MNF is as follows:

[0054] ;

[0055] in, is the frequency value of the myoelectric power spectrum within the frequency interval ( The frequency value within a frequency interval represents a frequency point in the myoelectric power spectrum). is the spectrum within the EMG power frequency interval ( The spectrum value within a frequency interval represents the energy or intensity of the signal at that frequency point). is the total number of frequency intervals.

[0056] The autoregressive (AR) model is a linear statistical model used to predict time series data. It assumes that the current observation depends only on past observations and not on future observations. The autoregressive model can be seen as a special deep learning model. The autoregressive model is implemented by recurrent neural networks (RNNs), and recurrent neural networks (RNNs) in deep learning can also be used to implement autoregressive models.

[0057] From the previous time-frequency analysis, we know that the sEMG signal has stationarity in the time domain and is a causal signal. Assuming that it is intercepted with a sufficiently short window length function, this sequence can be approximated as a linear sequence. When the muscle has a corresponding characteristic movement, it can be assumed that the sEMG in a certain data frame is a stable non-white noise linear sequence, which means that the sequence is a stable linear sequence containing relevant information. At this time, the AR value of the signal can be calculated as the feature of the signal.

[0058] The p-order AR model is defined as follows:

[0059] ;

[0060] in, Represents the observation value or output value of the time series at time t; is a constant term or intercept term, which means that when all lag terms are 0, Expected value; are autoregressive parameters, which constitute the parameter vector of the AR model. These parameters represent the linear relationship between the current observation and the past observation in the time series. Weighed Observed value at time The observed value at time t the extent of the impact; Indicates that the time series is The observed value at time These lagged terms are used as independent variables to predict or explain the observed values ​​at the current moment. ; is an error term or a random disturbance term, if is Gaussian noise, then it follows the standard normal distribution form.

[0061] It is worth noting that when the muscle fatigue state deepens, the RMS value generally increases, while the MPF shows a trend of first increasing and then decreasing. However, the regular changes of MPF are not consistent in different muscle groups. In linear complexity analysis, when the muscle is fatigued, the entropy representation of the surface electromyographic signal decreases, and the signal stability increases, that is, the stability of the AR model increases (the change of mean and variance over time decreases). In the subsequent muscle fatigue analysis, this application uses the stability of RMS and AR model as the main characteristic indicators, and MPF as an auxiliary detection indicator. In order to further study the relationship between muscle fatigue and maximum force, this application uses the normalized RMS of MVC (maximum value of sEMG signal under voluntary maximum contraction force) instead of RMS to reflect the difference between the maximum force during muscle fatigue and the maximum force in the non-fatigue state.

[0062] Specifically, see Figure 3 First, the maximum value of the surface electromyographic signal under the maximum voluntary contraction force (MVC) is measured, and the RMS normalized by MVC (i.e., MVC / RMS) is used as the new RMS evaluation. It should be noted that the normalized RMS reflects the difference between the maximum strength of the muscle group when the muscle is fatigued and the maximum strength of the muscle group when it is not fatigued, and can be used as a parameter for fatigue evaluation; the normalized RMS converts the numerical relationship into a ratio relationship, which is more conducive to eliminating individual differences between different individuals and improving accuracy.

[0063] It can be understood that to measure MVC: First, it is necessary to measure the maximum value of the surface electromyographic signal generated by the individual under the maximum voluntary contraction force. This value is called MVC (Maximum Voluntary Contraction). MVC represents the maximum electrical signal intensity that the individual can generate on a specific muscle group. Calculate RMS: During the experiment, it is necessary to continuously record and analyze the sEMG signal, and calculate the RMS value from it. RMS is a measure of signal amplitude that can reflect the intensity of muscle activity. Normalization: Next, the RMS is normalized using the measured MVC value. Specifically, the RMS value at each time point is divided by the MVC value to obtain the MVC / RMS ratio. This ratio reflects the ratio of the current RMS value to the MVC, that is, the ratio of the current muscle activity intensity to the maximum intensity. Assessing muscle fatigue: Normalized RMS (MVC / RMS) can be used as a parameter to assess muscle fatigue. During muscle fatigue, the RMS value usually increases with the increase in muscle activity, but since the MVC value represents the maximum intensity, the MVC / RMS ratio usually decreases during muscle fatigue. This means that as muscle fatigue deepens, the ratio of the current RMS value to the MVC decreases, reflecting the decrease in muscle activity intensity. Eliminating individual differences: An important advantage of normalization is that it can eliminate individual differences between different individuals. Since the MVC value represents the maximum intensity of each individual, the MVC / RMS ratio is more comparable between different individuals, which allows the normalized RMS to more accurately reflect the state of muscle fatigue without being affected by individual differences.

[0064] The present application can convert sEMG signals of different individuals into a relatively uniform standard through normalization processing, thereby improving the accuracy of assessing muscle fatigue.

[0065] In step S103, the robot arm obtains force information according to the fatigue index information, and applies force to the pulley assembly according to the force information to counteract the force of the user pulling the pull rope.

[0066] In a possible implementation, the force information includes a force magnitude and a force direction. The mechanical arm obtains the force magnitude and the force direction of the mechanical arm according to the time domain evaluation parameter and the frequency domain evaluation parameter.

[0067] It should be noted that if Figure 3As shown, the electromyographic characteristic data includes RNS and MNF. The movement cycle of the robot arm is the same as the judgment cycle of the fatigue characteristic, both of which are large cycles T. If the subject is judged to be not fatigued, the robot arm rises (maintains no movement when it reaches the highest point), and the degree of mutual resistance between the arm and the robot arm increases. Otherwise, it falls (maintains no movement when it reaches the lowest point). It can be understood that the force applied or the moving distance of the robot arm is configurable, which determines the training intensity. Users can adjust the configuration according to their own situation.

[0068] This application ensures that joint stability during exercise is proven to be a key measure to effectively reduce the risk of sports injuries. Joint stability is a favorable guarantee for completing movements in accordance with standardized motion trajectories. The more accurate the movement, the less sports injuries and the better the training effect. In addition, joint stability and strength complement each other. When the muscle strength around the joint increases, the stability of the joint will increase. The better the stability of the joint, the more conditions can be provided for the development of muscle strength. This exercise method not only brings a new experience to traditional long contraction exercises such as plank support by emphasizing static force under joint stability, but also achieves the ultimate precise control of muscle stimulation with the help of the intelligent dynamic adaptive adjustment technology of the mechanical arm while maintaining the significant training effect of isometric contraction on muscle endurance and stability. With its advanced technology, the mechanical arm can accurately capture and adjust the force size and direction in real time according to the user's electromyographic signal. This design not only significantly improves the training effect, but also effectively avoids direct or cumulative injuries, muscle compensation, posture problems and prolonged recovery time caused by improper fitness, ensuring that users can enjoy efficient exercise while their sports safety is fully guaranteed.

[0069] In step S104, when the mechanical arm determines that the current state of the user is a fatigue state according to the fatigue index information, the mechanical arm releases the pulley assembly.

[0070] In a possible implementation, the robotic arm calculates instantaneous fatigue characteristics in adjacent preset time periods based on the fatigue index information; the robotic arm determines the instantaneous state of the user based on the instantaneous fatigue characteristics; within a preset period, if the robotic arm determines that the number of instantaneous states as instantaneous fatigue states is greater than a preset value, it determines that the current state of the user is a fatigue state.

[0071] For details, see Figure 4 , extract and calculate fatigue features in a short period of time t, and judge the fatigue status of the small period (fatigue in the instantaneous state) according to the features; the large period T is composed of n identical periods t; and the fatigue status of the large period T is evaluated according to the instantaneous fatigue status of n small periods t. (When at least n' of the n small periods are judged to be instantaneous fatigue, the large period T is considered to be in a fatigue state).

[0072] It can be understood that feature extraction and instantaneous fatigue judgment within a small cycle: within a short small cycle t, relevant fatigue features are extracted from the sEMG signal, such as RMS (root mean square value), MNF (mean frequency) or other indicators that can reflect the state of muscle activity. According to the change trend of these characteristic values ​​or the preset threshold, it is judged whether the muscle within the small cycle t is in an instantaneous fatigue state. The instantaneous fatigue state means that in this short period of time, the muscle activity intensity increases, the frequency distribution changes, etc., indicating that the muscle may be experiencing a fatigue process. Constructing a large cycle T: A large cycle T is composed of n identical small cycles t. This large cycle T can be a few minutes, hours or longer time period, depending on the experimental design and research purpose. Each small cycle t is independent, but they together constitute the overall data set within the large cycle T. Comprehensive evaluation within the large cycle: According to the instantaneous fatigue state within n small cycles t, the muscle fatigue within the large cycle T is comprehensively evaluated. The specific evaluation criteria can be: when at least n' (n' is a preset threshold, and n'≤n) of the n small cycles are judged to be instantaneous fatigue, the large cycle T is considered to be in a fatigue state. The n' value here can be adjusted according to the experimental design and research purpose to reflect different requirements for fatigue sensitivity. This application can take into account the cumulative effect of muscle fatigue at different time points, thereby more accurately evaluating the overall fatigue state within the large cycle T.

[0073] In a possible implementation, the instantaneous fatigue feature includes a first difference and a second difference. The robotic arm calculates the first difference of the time domain evaluation parameter in adjacent preset time periods and the second difference of the frequency domain evaluation parameter in adjacent preset time periods.

[0074] It can be understood that the first difference is the difference in RMS values ​​in adjacent preset time periods (ie, in a small period t), and the second difference is the difference in RMS values ​​in adjacent preset time periods.

[0075] In one possible implementation, if the robotic arm determines that the first difference and the second difference are both greater than zero, it determines that the user's instantaneous state is an instantaneous fatigue state; if the robotic arm determines that the first difference is less than or equal to zero, it determines that the user's instantaneous state is an instantaneous non-fatigue state.

[0076] Furthermore, in the instantaneous situation, it is judged based on the time domain feature RMS (root mean square value) and the frequency domain feature MNF (mean frequency); in the muscle groups in the experiment, when the muscle fatigue state deepens, the RMS value generally increases, while the MPF shows a trend of first increasing and then decreasing; the difference between adjacent RMS and MNF under the small cycle t is calculated, and if it is greater than 0, it is considered that the value is increasing in the instantaneous situation, otherwise it is decreasing; the fatigue situation of the instantaneous situation is judged based on the fatigue trend.

[0077] It can be understood that in the instantaneous case, the time domain feature RMS and the frequency domain feature MNF are used to judge: at a specific time point or time window (instantaneous case), by calculating the RMS and MNF values ​​of the sEMG signal, the fatigue state of the muscle can be preliminarily judged. RMS reflects the amplitude change of the signal and is related to the intensity of muscle activity; while MNF reflects the characteristics of the signal in the frequency domain and is related to the frequency distribution of muscle activity. In the muscle group in the experiment, when the muscle fatigue state deepens, the RMS value generally increases, while the MPF shows a trend of increasing first and then decreasing. Calculate the difference between adjacent RMS and MNF under a small cycle t. If it is greater than 0, it is considered that the value is increasing in the instantaneous case, otherwise it is decreasing: in a shorter time period (small cycle t), calculate the difference between RMS and MNF of adjacent time points or data points. If the difference is greater than 0, it means that the characteristic value is increasing in the instantaneous case; if the difference is less than 0, it means that the characteristic value is decreasing. The purpose of this step is to observe the changing trend of RMS and MNF in a short time more finely. Judging the instantaneous fatigue situation based on fatigue trends: Combine the changing trends of RMS and MNF, as well as the calculated difference, to judge the fatigue state of the muscle in an instantaneous situation. Specifically, if the RMS value is increasing, while the MNF value is decreasing (or first increasing and then decreasing to below the initial value), this may indicate that the muscle is undergoing a fatigue process. However, it should be noted that due to differences between different muscle groups and individuals, the changing trends of these characteristic values ​​may be different. Therefore, in practical applications, it may be necessary to combine multiple characteristic indicators and the specific situation of the individual to make a comprehensive judgment.

[0078] This application can identify relevant information about the user's movement by collecting the human body's surface electromyographic signals and performing real-time analysis, and then control the movement of the robotic arm to achieve personalized control of the movement to meet the exercise needs of different users, and through real-time feedback, help users understand whether their muscle activity meets the expected exercise goals, and achieve good exercise results through deep perception of the human body state and intelligent interaction. Real-time monitoring and analysis of electromyographic signals can determine the current state of movement and fatigue level, and can effectively prevent fatigue injuries caused by excessive exercise. Near-static movement patterns can reduce muscle tissue wear, improve stability, and reduce the probability of sports injuries.

[0079] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0080] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0081] It should be understood that the application of the present application is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to this application.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A control method for a closed-loop human-computer interaction device, characterized in that: The closed-loop human-computer interaction device comprises a pull rope, a pulley assembly, a measuring assembly and a mechanical arm, wherein the pull rope, the pulley assembly and the mechanical arm are connected in sequence, and the measuring assembly is connected to the mechanical arm; The closed-loop human-computer interaction device comprises a pull rope, a pulley assembly, a measuring assembly and a mechanical arm, wherein the pull rope, the pulley assembly and the mechanical arm are connected in sequence, and the measuring assembly is connected to the mechanical arm; The pulley assembly comprises a fixed plate, a slide plate, a connecting rope, a first pulley, a second pulley, a third pulley and a fixed wheel, the slide plate is slidably connected to the fixed plate, the first pulley and the third pulley are connected to the slide plate, the second pulley and the fixed wheel are connected to the fixed plate, the fixed wheel is located on a side of the fixed plate away from the second pulley, one end of the connecting rope is fixed to a side close to the second pulley, and the connecting rope is connected to the mechanical arm through the third pulley, the second pulley and the first pulley in sequence, one end of the pull rope is fixed to the slide plate, and the pull rope is connected to the fixed wheel, and the other end of the pull rope is used to be pulled by a user; When the user pulls the pull rope, the mechanical arm applies force to the pulley assembly to counteract the force of the user pulling the pull rope; The control method of the closed-loop human-computer interaction device comprises: When the user pulls the pull rope, the measuring component obtains the user's electromyographic information; The measuring component obtains fatigue index information according to the electromyographic information and sends the information to the robotic arm; The mechanical arm obtains force information according to the fatigue index information, and applies force to the pulley assembly according to the force information so as to counteract the force of the user pulling the pull rope; When the mechanical arm determines that the current state of the user is a fatigue state according to the fatigue index information, the mechanical arm releases the pulley assembly; The electromyographic information includes target electromyographic signals, real-time electromyographic signals and frequency data; When the user pulls the pull rope, the measuring component obtains the user's electromyographic information, specifically including: When the user pulls the pull rope with a target force, the measurement component obtains the user's target electromyographic signal; When the user pulls the pull rope with the current force, the measuring component obtains the user's real-time electromyographic signal and real-time frequency data; The fatigue index information includes time domain evaluation parameters and frequency domain evaluation parameters; The measuring component obtains fatigue index information according to the electromyographic information, specifically including: The measuring component calculates a root mean square value according to the real-time electromyographic signal, and obtains a frequency domain evaluation parameter according to the real-time frequency data; The measuring component obtains a time domain evaluation parameter according to the target electromyographic signal and the root mean square value; The measuring component calculates the root mean square value according to the real-time electromyographic signal, and obtains the frequency domain evaluation parameter according to the real-time frequency data, specifically including: The measuring component calculates a mean square root according to the real-time electromyographic signal, and normalizes the mean square root according to the target electromyographic signal to obtain a time domain evaluation parameter; The calculation formula of the root mean square value is as follows: ； in, is the root mean square value, Indicates the length of the time window, indicating the time from the starting point ( ) to the current point ( ) time interval; represents the integral variable; Indicates the current time or the end point of the analyzed time window, is the surface electromyographic signal; use As the normalized ,in, represents the maximum value of the sEMG signal under the maximum voluntary contraction force, using After normalization Come as a new Evaluate; The mechanical arm determines, according to the fatigue index information, that the current state of the user is a fatigue state, specifically including: The robotic arm calculates instantaneous fatigue characteristics in adjacent preset time periods according to the fatigue index information; The mechanical arm determines the instantaneous state of the user according to the instantaneous fatigue characteristic; If, within a preset period, the number of times that the robot arm determines that the instantaneous state is an instantaneous fatigue state is greater than a preset value, it is determined that the current state of the user is a fatigue state.

2. The control method of the closed-loop human-computer interaction device according to claim 1, characterized in that: The force information includes the force magnitude and the force direction; The mechanical arm obtains force information according to the fatigue index information, specifically: The robotic arm obtains the force magnitude and force direction of the robotic arm according to the time domain evaluation parameters and the frequency domain evaluation parameters.

3. The control method of the closed-loop human-computer interaction device according to claim 1, characterized in that: The instantaneous fatigue characteristic includes a first difference and a second difference; The robotic arm calculates instantaneous fatigue characteristics in adjacent preset time periods according to the fatigue index information, specifically: The robotic arm calculates a first difference value of the time domain evaluation parameter in adjacent preset time periods and a second difference value of the frequency domain evaluation parameter in adjacent preset time periods.

4. The control method of the closed-loop human-computer interaction device according to claim 3, characterized in that: The mechanical arm determines the instantaneous state of the user according to the instantaneous fatigue characteristic, specifically including: If the mechanical arm determines that both the first difference and the second difference are greater than zero, it determines that the instantaneous state of the user is an instantaneous fatigue state; If the robotic arm determines that the first difference is less than or equal to zero, it determines that the instantaneous state of the user is an instantaneous non-fatigue state.

5. A closed-loop human-computer interaction device for implementing the control method of the closed-loop human-computer interaction device according to any one of claims 1 to 4, characterized in that: The closed-loop human-computer interaction device comprises a pull rope, a pulley assembly, a measuring assembly and a mechanical arm, wherein the pull rope, the pulley assembly and the mechanical arm are connected in sequence, and the measuring assembly is connected to the mechanical arm; The pulley assembly comprises a fixed plate, a slide plate, a connecting rope, a first pulley, a second pulley, a third pulley and a fixed wheel, the slide plate is slidably connected to the fixed plate, the first pulley and the third pulley are connected to the slide plate, the second pulley and the fixed wheel are connected to the fixed plate, the fixed wheel is located on a side of the fixed plate away from the second pulley, one end of the connecting rope is fixed to a side close to the second pulley, and the connecting rope is connected to the mechanical arm through the third pulley, the second pulley and the first pulley in sequence, one end of the pull rope is fixed to the slide plate, and the pull rope is connected to the fixed wheel, and the other end of the pull rope is used to be pulled by a user; When the user pulls the pull rope, the mechanical arm applies force to the pulley assembly to counteract the force of the user pulling the pull rope.

6. The closed-loop human-computer interaction device according to claim 5, characterized in that: The closed-loop human-computer interaction device also includes a breathing light, the measuring component is connected to the breathing light, and the electromyographic information collected by the measuring component is sent to the breathing light, so that the breathing light flashes according to the electromyographic information.

Images