Acupuncture point massage strength control method for intelligent moxibustion robot

Abstract

The invention discloses an acupoint massage strength control method for an intelligent moxibustion robot, and relates to the technical field of intelligent medical equipment. The method comprises the following steps: acquiring human body acupuncture point positioning data and real-time contact force data through a visual sensor and a force sensor; a force control model is constructed based on the acupoint type, the human physiological characteristics and the moxibustion temperature parameters; an impedance control algorithm and a dynamic threshold adjustment strategy are combined to realize self-adaptive adjustment of massage strength; cooperative parameters of the acting force of the massage head and the moxibustion temperature are corrected in real time through a closed-loop feedback mechanism. The problems that in the prior art, the adaptability of massage strength, acupoint characteristics and human body differences is poor, and moxibustion and massage cooperative control precision is insufficient are solved, precise and personalized control over the acupoint massage strength is achieved, the moxibustion and massage cooperative physiotherapy effect is improved, and meanwhile safety and comfort in the physiotherapy process are guaranteed.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical device technology, specifically to a method for controlling the intensity of acupoint massage in an intelligent moxibustion robot. Background Technology

[0002] With the development of the health industry, the combination of moxibustion and massage has gained widespread attention, leading to the emergence of intelligent moxibustion robots. However, current technologies for controlling the massage intensity of moxibustion robots mostly employ fixed parameter modes or simple force feedback adjustments, which present the following technical problems: The massage intensity is poorly matched with the characteristics of acupoints. Different types of acupoints (such as core acupoints and peripheral acupoints) have different intensity requirements, but existing methods do not make targeted distinctions, resulting in poor therapeutic effects. The massage intensity may not fully take into account the differences in human physiological characteristics, such as weight, muscle thickness, and skin elasticity, which can lead to discomfort due to excessive intensity or no effect due to insufficient intensity. The lack of coordinated control between moxibustion temperature and massage intensity means that the synergistic effect between the two is not effectively utilized, thus affecting the therapeutic effect. The response speed of the intensity adjustment is slow, the feedback mechanism is not sound, and it is difficult to adapt to the dynamic changes in human posture and the sensitivity of acupoints in real time. The lack of effective collision detection and safety protection mechanisms means that when the human body moves accidentally or the massage head deviates from its positioning, excessive local force can easily occur, posing a safety hazard.

[0003] These issues result in insufficient precision and poor personalization of existing intelligent moxibustion robots in acupoint massage, making it difficult to meet users' dual needs for therapeutic effects and safety, thus limiting their promotion and application in the healthcare field. Summary of the Invention

[0004] In view of this, the purpose of this invention is to overcome the shortcomings of the prior art and provide a method for controlling the intensity of acupoint massage in an intelligent moxibustion robot. Through multi-sensor data fusion, intelligent algorithm modeling, and closed-loop feedback adjustment, the method achieves precise and personalized control of massage intensity, while taking into account the synergistic effect of moxibustion and massage and the safety of physiotherapy.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: The first objective of this invention is to provide a method for controlling the intensity of acupoint massage in an intelligent moxibustion robot, comprising the following steps: S1: Collect human images and physiological feature data of the target area, locate the acupoints and their types through image recognition algorithms, and obtain acupoint location information; S2: Based on the acupoint positioning information, control the massage head of the intelligent moxibustion robot to move to the target acupoint, and collect the contact force data between the massage head and the human body in real time through the force sensor; S3: Obtain the real-time working temperature data of the moxibustion component, and construct a force control model by combining the acupoint positioning information, physiological characteristic data and contact force data; S4: Based on the force control model, the target massage force parameters are calculated using an impedance control algorithm; S5: Control the massage head to output corresponding force according to the target massage force parameter, and adjust the massage force and moxibustion temperature in real time through a closed-loop feedback mechanism to achieve coordinated control.

[0006] Furthermore, the physiological characteristic data includes body weight, muscle thickness, skin elasticity, and acupoint sensitivity coefficient, which is obtained through a preset exploratory massage process.

[0007] Furthermore, the acupoint location information includes acupoint coordinates, the meridian to which the acupoint belongs, and the acupoint type. The acupoint type includes core acupoints, peripheral acupoints, and pain point acupoints.

[0008] Furthermore, the construction process of the force control model includes: The basic pressure range is determined based on the type of acupoint, with the basic pressure range for core acupoints being higher than that for peripheral acupoints. The basic force range is corrected by combining physiological characteristic data; the greater the muscle thickness, the greater the force correction coefficient. Introducing the temperature-influencing factor of moxibustion, when the moxibustion temperature is in the range of 38-43℃, the intensity adjustment coefficient increases by 5%-8% for every 1℃ increase in temperature.

[0009] Furthermore, the impedance control algorithm adopts the following formula: ΔF = K·ΔX + B·ΔV Where ΔF is the force adjustment amount, K is the dynamic stiffness coefficient, ΔX is the massage head position offset amount, B is the damping coefficient, and ΔV is the change in the massage head movement speed.

[0010] Furthermore, the dynamic stiffness coefficient K The adaptive adjustment method is as follows: When the actual contact force is less than the maximum permissible contact force

[0011] When the actual contact force is greater than or equal to the maximum permissible contact force ; in, For the maximum permissible contact force, For actual contact force, This is the initial stiffness coefficient.

[0012] Furthermore, the closed-loop feedback mechanism includes: Real-time acquisition of contact force data and moxibustion temperature data; calculation of the deviation between actual and target values. When the deviation exceeds the preset threshold, adjust the force of the massage head and the heating power of the moxibustion component; Feedback adjustment is completed every 50-100ms to ensure that the force deviation does not exceed ±0.5N and the temperature deviation does not exceed ±0.3℃.

[0013] Furthermore, it also includes a collision detection step: a torque observer monitors sudden changes in the force on the massage head. When the detected torque change exceeds a preset threshold, the massage head is controlled to immediately reduce the force and adjust its position to avoid injury to the human body.

[0014] Furthermore, the massage head 1 includes an EMS massage head, the pulse current intensity of which is linked to the massage force parameter for control. The greater the force, the higher the corresponding pulse current intensity level, and the pulse current intensity is adjustable in 3-5 levels.

[0015] Furthermore, it also includes a physiotherapy mode switching step: automatically switching between acupoint opening mode, strengthening mode and acupoint sealing mode based on the user's physiotherapy stage. The target massage intensity, moxibustion temperature and feedback adjustment frequency are different in different modes. The acupoint opening mode is gentle, the strengthening mode is strong, and the acupoint sealing mode tends to be gentle.

[0016] Compared with the prior art, the present invention has at least the following beneficial effects: This acupoint massage intensity control method combines acupoint type, human physiological characteristics, and moxibustion temperature to construct an intensity control model, providing precise intensity parameters for different users and acupoints to enhance therapeutic effects. Moxibustion temperature is used as a key influencing factor in intensity adjustment, achieving a synergistic effect that enhances meridian unblocking and metabolism promotion. Employing an impedance control algorithm and a high-frequency closed-loop feedback mechanism, the intensity deviation is no more than ±0.5N, and the temperature deviation is no more than ±0.3℃, adapting in real-time to changes in body posture and sensitivity. It boasts high safety through multiple safety protection mechanisms, including sudden force detection, maximum intensity limitation, and temperature protection, preventing injury during therapy. It supports mode switching between different therapy stages, meeting users' full-cycle therapy needs from acupoint opening and strengthening to acupoint sealing, while also being compatible with users of different weights and muscle thicknesses. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the acupoint massage intensity control method for an intelligent moxibustion robot in an embodiment of the present invention. Figure 2This is a schematic diagram illustrating the actual use of the intelligent moxibustion robot performing acupoint massage in an embodiment of the present invention; Figure 3 This is a frontal schematic diagram of the distribution of human acupoint locations and types in an embodiment of the present invention; Figure 4 This is a side view of the distribution of acupoints and their types in an embodiment of the present invention. Figure 5 This is a schematic diagram of the specific process of step S3 in an embodiment of the present invention; Figure 6 This is a three-dimensional structural diagram of the intelligent moxibustion robot in an embodiment of the present invention.

[0018] Explanation of reference numerals in the attached figures: 1-Vision sensor; 2-Force sensor; 3-Robotic arm. Detailed Implementation

[0019] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In the description of this invention, it should be noted that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can also refer to the internal connection of two components; and they can refer to a wireless connection or a wired connection. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0020] Please see Figure 1 As shown, this embodiment of the invention provides a method for controlling the intensity of acupoint massage in an intelligent moxibustion robot. The method includes the following steps: S1: Collect human images and physiological feature data of the target area, locate the acupoints and their types through image recognition algorithms, and obtain acupoint location information.

[0021] In this step, basic data collection and precise positioning are the prerequisites for the entire control process. Traditional manual moxibustion and massage rely on experience to locate acupoints, which has a large error. However, this step collects human images and physiological feature data, and uses image recognition algorithms to locate acupoints. This can digitize the location and type of acupoints, providing a coordinate reference for the precise implementation of subsequent massage and moxibustion. For example, it can distinguish acupoints with different effects, such as Zusanli and Guanyuan, and avoid deviations in the application of force and the application site.

[0022] S2: Based on the acupoint positioning information, control the massage head 1 of the intelligent moxibustion robot to move to the target acupoint, and collect the contact force data between the massage head 1 and the human body in real time through the force sensor 2.

[0023] In this step, the connection between mechanical execution and force data acquisition is completed by taking over the positioning results from S1. After the massage head 1 is precisely moved to the target acupoint by the robotic arm 3, the force sensor 2 collects contact force data in real time. This step can capture the basic force situation of the massage in real time. For example, the initial contact force of acupoints with thicker muscles may be larger, while the initial contact force of acupoints with thinner muscles may be smaller, providing real-time force feedback data for subsequent intensity adjustment.

[0024] S3: Obtain the real-time operating temperature data of the moxibustion component, and construct a force control model by combining the acupoint positioning information, physiological characteristic data and contact force data.

[0025] In this step, the warming effect of moxibustion affects the condition of the skin and muscles, thereby altering the body's tolerance to massage pressure. For example, muscles are more relaxed at higher temperatures, and the tolerance to pressure may be higher. This step integrates key data such as moxibustion temperature, acupoint information, physiological characteristics, and contact force to construct a pressure control model that can overcome the limitations of single force data, allowing pressure control to conform to the actual scenario of the synergistic effect of moxibustion and massage.

[0026] S4: Based on the force control model, the target massage force parameters are calculated using an impedance control algorithm.

[0027] Step S4 is the core computational step for force control. Impedance control algorithms are commonly used for force control in robot-human contact. By simulating the impedance characteristics of human skin and muscles, appropriate force parameters can be calculated. Based on the model in step S3, this algorithm can transform multi-source data into executable target massage force parameters. For example, it can calculate differentiated force thresholds for elderly people with weak constitutions and young people with healthy constitutions, avoiding ineffective force due to insufficient force or injury due to excessive force.

[0028] S5: Control the massage head 1 to output the corresponding force according to the target massage force parameter, and adjust the massage force and moxibustion temperature in real time through a closed-loop feedback mechanism to achieve coordinated control.

[0029] In this step, after outputting the target intensity, it is not executed statically, but rather the massage intensity and moxibustion temperature are continuously monitored through a closed-loop feedback mechanism. For example, if the feedback shows that the contact force exceeds the human body's tolerance threshold, the system can immediately reduce the massage intensity; if the moxibustion temperature is too high and causes the skin to burn, the massage intensity can be finely adjusted while lowering the temperature, achieving synergistic adaptation between the two and solving the problem of the disconnect between massage intensity and moxibustion temperature in traditional methods.

[0030] In traditional manual massage, the massage intensity and moxibustion temperature rely entirely on the technician's experience, leading to significant variations in technique between different technicians or even the same technician at different times. This new solution, however, uses image recognition to locate acupoints, greatly reducing errors. By combining a model built from force sensor 2 and temperature data with an impedance control algorithm to calculate intensity parameters, the massage intensity error can be controlled to an extremely small range. Simultaneously, the coordinated control of moxibustion temperature and massage intensity ensures consistent parameters for each treatment. For example, when targeting the Shenshu acupoint for patients with lumbar strain, the appropriate intensity and temperature can be precisely applied each time, guaranteeing the stability of the therapeutic effect.

[0031] Furthermore, the force and heat tolerance thresholds vary greatly among different acupoints and different populations. For example, children, the elderly, and young adults have different tolerance levels, and the upper limit of force applied to acupoints near joints differs from that applied to acupoints in areas with abundant muscle. In this solution, physiological characteristic data reflects differences in user constitution, and force sensor 2 and temperature data provide real-time feedback. The closed-loop feedback mechanism can quickly respond to abnormalities. Once the contact force exceeds the limit or the temperature is too high, the system immediately adjusts, effectively avoiding muscle damage caused by massage or burns caused by moxibustion, thus solving the pain point of difficulty in real-time control of safety hazards in traditional moxibustion massage.

[0032] Finally, the warming effect of moxibustion and the mechanical stimulation of massage complement each other in the treatment process. Moxibustion can warm and unblock the meridians, while massage can relax muscles; the synergy of the two can improve the efficiency of the treatment. The intensity control model of this program integrates multi-dimensional data and can customize programs for different users. For example, for patients with dysmenorrhea, gentle massage intensity is applied to the Guanyuan and Sanyinjiao acupoints, combined with appropriate moxibustion temperature. This can relieve muscle spasms through massage and warm the meridians and dispel cold through moxibustion, achieving personalized synergistic treatment and improving the treatment effect.

[0033] This solution replaces the experience-based judgment and real-time monitoring required for manual operation with algorithms and sensors. For example, technicians no longer need to constantly adjust the massage intensity and moxibustion temperature; the robot can autonomously complete data collection, parameter calculation, and dynamic adjustments. This allows technicians to manage multiple devices simultaneously, significantly increasing service capacity and reducing labor costs. Furthermore, the standardized process makes the intelligent moxibustion robot more suitable for diverse scenarios such as hospitals, health centers, and homes.

[0034] More specifically, in one embodiment of the present invention, the physiological characteristic data includes body weight, muscle thickness, skin elasticity, and acupoint sensitivity coefficient, wherein the acupoint sensitivity coefficient is obtained through a preset exploratory massage process.

[0035] This technical solution further improves the precision and personalization of force control by clarifying the specific dimensions of physiological characteristic data and supplementing the collection method of acupoint sensitivity coefficients. The newly added physiological characteristic data dimensions and sensitivity coefficient collection methods are more in line with the actual physiological differences of the human body, and each data point plays a clear technical role, as follows: Body weight and muscle thickness are the basic physical basis for adapting massage pressure. Body weight and muscle mass are usually positively correlated, and muscle thickness directly determines the degree of pressure buffering at acupoints. For example, users with greater weight and thicker muscles need greater initial pressure to achieve effective massage depth; users with less weight and thinner muscles (such as children and the elderly) need less initial pressure to avoid injury. These two data provide a basic force threshold reference for the pressure control model.

[0036] Skin elasticity reflects the resilience and recovery ability of human tissues, and is directly related to the tolerance of massage pressure. Users with better skin elasticity (such as young people) are more adaptable to changes in pressure and can appropriately expand the range of pressure adjustment; users with poor skin elasticity (such as the elderly and those with weak constitutions) have insufficient tissue resilience and need to control the amplitude of pressure fluctuations to avoid damage to the skin or subcutaneous tissue. This data supplements the model with a dynamic force adaptation dimension.

[0037] Acupoint sensitivity coefficient is the core data for personalized intensity control. Different users have vastly different sensitivities to the same acupoint. Even with similar weight and muscle thickness, users with high sensitivity coefficients may react noticeably to slight pressure, while users with low sensitivity coefficients require greater pressure to perceive any effect. By collecting this data through preset trial massages, it is possible to avoid blindly setting the intensity. For example, by testing the acupoint response with gradually increasing slight pressure, the range of pressure that elicits a comfortable feedback from the user can be recorded, transforming subjective feelings into objective data and providing personalized comfort thresholds for the model.

[0038] The preliminary data collection process for the trial massage provides a pre-calibration function for the subsequent formal massage. This process is completed before the formal massage, without affecting the core treatment process, but it allows the intensity control model to obtain the user's personalized data in advance, avoiding intensity adaptation errors in the early stages of the formal massage. This makes the initial output target intensity closer to the user's needs, reducing the adjustment range and time of closed-loop feedback.

[0039] The addition of physiological data makes the adjustment of closed-loop feedback more targeted. For example, if the closed-loop feedback finds that the contact force does not meet the expected effect, and it is determined by combining muscle thickness data that the muscle is too thick and the force is not penetrating enough, the force can be increased directly; if it is determined by combining acupoint sensitivity coefficient that the user's sensitivity is too low, the moxibustion temperature can be finely adjusted while increasing the force, and the heat effect can be used to help improve perception, avoiding the discomfort that may be caused by simply adjusting the force.

[0040] Therefore, by refining the dimensions of physiological characteristic data and clarifying the collection methods for sensitivity coefficients, the model achieves more significant results in terms of accuracy, personalization, and safety. Objective data such as weight and muscle thickness provide quantitative references for the intensity applied, ensuring that the target intensity calculated by the model no longer relies on vague group averages. For example, for the same acupoint, the model will calculate significantly different intensity parameters for users with a muscle thickness of 1.5cm and those with a muscle thickness of 3cm, ensuring that the former is not subjected to excessive force and the latter achieves an effective massage depth, thus improving the stability of the therapeutic effect.

[0041] For users with different sensitivities, the system collects acupoint sensitivity coefficients. Sensitive users can be identified through trial massage, and the model will output gentle and stable pressure. Users with high tolerance will receive a pressure more closely matched to their perception, avoiding ineffective treatment due to insufficient pressure. Furthermore, the inclusion of data such as skin elasticity and weight allows the system to be adapted to different groups, including children, the elderly, young adults, obese individuals, and underweight individuals, expanding the applicability of the intelligent moxibustion robot.

[0042] For example, for elderly people with poor skin elasticity, the model limits the maximum pressure and the rate of pressure adjustment; for children with weak muscles, it sets an extremely low initial pressure based on weight data, and then adjusts it through trial massage, providing double protection to avoid injury. Compared to solutions that rely solely on contact force data, the addition of data can predict risks in advance, improving the safety of the operation. The trial massage process makes the initial pressure of the formal massage more in line with the user's body sensation, avoiding discomfort such as stinging and soreness caused by improper initial pressure. At the same time, the model combines sensitivity coefficient and skin elasticity data to optimize the smoothness of pressure changes. For example, when the pressure increases, users with high sensitivity coefficients have a slower rate of increase, reducing the impact on the body and improving the comfort of the massage process.

[0043] For more details, please refer to Figure 2 , 3 As shown in Figure 4, in one embodiment of the present invention, the acupoint location information includes acupoint coordinates, the meridian to which the acupoint belongs, and the acupoint type. The acupoint type includes core acupoints, peripheral acupoints, and pain point acupoints.

[0044] In this embodiment, by accurately locating the acupoint coordinates, the massage head 1 can achieve millimeter-level movement and positioning, avoiding the estimation deviation of the approximate area of ​​the acupoint in traditional positioning, ensuring that the massage force and moxibustion heat are accurately applied to the target acupoint, and providing a more precise range of action for subsequent force and temperature control, such as preventing the force of the core acupoint from shifting to the surrounding non-acupoint area.

[0045] Different meridians correspond to different organ functions, such as the Foot Taiyin Spleen Meridian related to digestion and the Hand Shaoyin Heart Meridian related to heart function. Meridian properties affect the body's tolerance and response to pressure and temperature. For example, when treating Heart Meridian acupoints, excessive pressure should be avoided to prevent stimulating the mind, while pressure can be appropriately increased when treating Bladder Meridian acupoints (located in areas with abundant muscle). Core acupoints (such as Guanyuan and Zusanli) are the core targets for treatment, requiring priority to ensure precise and stable pressure. Peripheral acupoints (such as auxiliary acupoints on meridian branches) can be used for supplementary treatment, with pressure slightly lower than that of core acupoints. Pain point acupoints (such as Ashi points caused by strain) usually involve muscle tension or inflammation, requiring gradient pressure to avoid direct, strong stimulation that could cause discomfort. This distinction between these three types of acupoints shifts pressure control from a uniform standard to a categorized and adaptive approach.

[0046] For different users, the collected data includes combinations such as core acupoints, thick muscles, and low sensitivity coefficients. The model will calculate a higher and more stable intensity. For combinations such as pain points, poor skin elasticity, and high sensitivity coefficients, the intensity will be output in a gentle and progressive manner, while a slightly lower moxibustion temperature will be used to reduce stimulation.

[0047] For example, when massaging core acupoints, if the feedback shows that the contact force is sufficient but the user's physiological characteristics (such as heart rate) fluctuate, the force can be reduced in combination with the meridian attributes (such as the Heart Meridian); when massaging painful acupoints, if the temperature rises and the skin's tolerance decreases, the temperature can be reduced at the same time while maintaining the current force to avoid double stimulation.

[0048] Therefore, by clearly defining the coordinates of acupoints, the error in the application point of the massage head 1 and the moxibustion component is significantly reduced, avoiding the problem of inaccuracy and ineffectiveness. For example, when the diameter of an acupoint is only a few millimeters, three-dimensional coordinate positioning can ensure that the force is concentrated in the core area of ​​the acupoint, improving the penetration of the massage and the heat conduction efficiency of the moxibustion.

[0049] By incorporating meridian attributes, the treatment becomes more aligned with Traditional Chinese Medicine (TCM) theory, avoiding the deviation in efficacy caused by targeting the right acupoints but accessing the wrong meridians. For example, when treating digestive issues, precise application to core acupoints on the spleen meridian, combined with appropriate pressure and temperature, can enhance the targeted effects of strengthening the spleen and stomach, rather than simply providing physical stimulation.

[0050] Core acupoints serve as the primary targets, and stable and precise pressure ensures adequate stimulation depth. For example, Zusanli (ST36), a core acupoint for health maintenance, benefits from consistent and stable pressure to better regulate Qi and blood. The auxiliary pressure applied to peripheral acupoints avoids wasting resources. Peripheral acupoints do not require high-intensity stimulation; moderate pressure assists the core acupoints while reducing unnecessary energy consumption and discomfort. Gradual pressure applied to pain points reduces the risk of stimulation. Pain points often present with inflammation or muscle tension; gradually increasing pressure allows the muscles to adapt, preventing exacerbation of pain from direct, strong stimulation. Combined with gentle moxibustion heat, the warmth can relieve tension and improve comfort.

[0051] For more details, please refer to Figure 5 As shown, in step S3, the process of constructing the force control model includes: Step S 31 The basic force range is determined based on the type of acupoint, with the basic force range for core acupoints being higher than that for peripheral acupoints. Preferably, in this embodiment, the basic force range for core acupoints is 15-30N, the basic force range for peripheral acupoints is 8-18N, and the basic force range for pain point acupoints is 12-25N.

[0052] Step S 32 The basic force range is corrected by combining physiological characteristic data; the greater the muscle thickness, the greater the force correction coefficient. Preferably, in this embodiment, the force correction coefficient increases by 10%-15% for every 1cm increase in muscle thickness; and the force correction coefficient decreases by 8%-10% for every 0.1 decrease in skin elasticity coefficient.

[0053] Step S 33 This study introduces a temperature-influencing factor for moxibustion. When the moxibustion temperature is within the range of 38-43℃, the intensity adjustment coefficient increases by 5%-8% for every 1℃ increase in temperature. This temperature range represents the comfortable and tolerable temperature range for moxibustion. Higher temperatures relax muscles and accelerate blood circulation, allowing for a more effective transmission of massage stimulation, thus achieving a synergistic effect between warmth and intensity. When the temperature exceeds 43℃, the intensity adjustment coefficient no longer increases, and a temperature protection mechanism is activated to prevent damage caused by the combination of high temperature and high intensity.

[0054] Finally, by integrating the above parameters, a multi-dimensional intensity control model is constructed, and the target massage intensity parameters are output.

[0055] Therefore, the model first uses acupoint type as the core basis for intensity allocation. Core acupoints are key targets for treatment and require sufficient intensity to ensure stimulation depth, so a higher base range of 15-30N is set. Peripheral acupoints play only an auxiliary role, and a range of 8-18N can avoid overstimulation. Pain point acupoints, due to muscle tension or inflammation, are given a middle range of 12-25N, which balances stimulation effect and leaves a safety margin for subsequent adjustments. This transforms intensity control from a vague judgment into a quantitative standard, avoiding arbitrariness in the base intensity under different scenarios and providing a unified benchmark framework for subsequent adjustments. For example, the core acupoint massage for all users initially falls within the 15-30N range, and then adjustments are made according to individual differences.

[0056] Based on the established range of pressure applied to acupoints, personalized adjustments are made in conjunction with physiological characteristic data: muscle thickness serves as the core adjustment factor, and its positive correlation with the pressure adjustment coefficient aligns with the laws of physical force application. Increased muscle thickness enhances force buffering, resulting in a decrease in stimulation depth at the same pressure. Each 1cm increase in muscle thickness elevates the adjustment coefficient by 10%-15%, ensuring that the pressure penetrates deep into the acupoints and preventing ineffective massage due to thick muscles.

[0057] The skin elasticity is negatively correlated with the force correction coefficient, and the setting is designed to fit the characteristics of human tissue. Reduced skin elasticity means decreased tissue toughness and tolerance. For every 0.1 decrease, the correction coefficient is lowered by 8%-10%, which can avoid damage to fragile tissues caused by excessive force, making it especially suitable for the elderly, those with weak constitutions, and other groups.

[0058] By quantifying objective data such as weight and muscle thickness, the base force range can be accurately matched to users with different body types. For example, for users with thick muscles and good skin elasticity, the adjusted force will be close to the upper limit of the base range; for users with thin muscles and poor skin elasticity, the adjusted force will be biased towards the lower limit of the base range.

[0059] More specifically, in one embodiment of the present invention, the impedance control algorithm employs the following formula: ΔF = K·ΔX + B·ΔV Where ΔF is the force adjustment amount, K is the dynamic stiffness coefficient, ΔX is the position offset of massage head 1, B is the damping coefficient, and ΔV is the change in the moving speed of massage head 1.

[0060] This technical solution, by clearly defining the quantitative formula of the impedance control algorithm, directly correlates the positional offset and speed change of the massage head 1 with the force adjustment. This design of the impedance control algorithm formula better reflects the dynamic characteristics of robot-human contact, enabling real-time dynamic correction of the force. Specifically: ΔF (force adjustment amount) is the algorithm's output target, directly corresponding to the massage force value that needs to be adjusted in the closed-loop feedback. This parameter inherits the target force output by the force control model mentioned earlier. When there is a deviation between the actual contact state and the expectation, ΔF is used to achieve real-time fine-tuning of the force, ensuring that the force always matches the target parameter.

[0061] K (dynamic stiffness coefficient) is the core parameter relating positional offset and force adjustment. ΔX (positional offset of massage head 1) reflects the deviation between the actual position of massage head 1 and the coordinates of the target acupoint. For example, if muscle deformation causes massage head 1 to offset from the target position by 1mm, the K value will convert this positional deviation into corresponding force compensation. The dynamic characteristics of the K value can be adapted to different acupoints and physiological states. For example, the K value is higher for core acupoints to ensure rapid force compensation when the positional offset occurs, preventing stimulation from deviating from the acupoint.

[0062] B (damping coefficient) is the key buffer for speed changes and intensity adjustments. ΔV (change in the speed of massage head 1) reflects fluctuations in the movement rate of massage head 1. For example, when contacting areas of thick muscle, the speed slows down, and the B value adjusts the intensity according to the change in speed, avoiding the impact of sudden speed changes. The introduction of the damping coefficient makes intensity adjustments smoother and reduces abruptness in the sensation.

[0063] It is understandable that the state of human muscles and acupoints is not fixed. During massage, muscle relaxation, breathing fluctuations, etc., can cause the massage head 1 to shift position or change speed. The algorithm quickly calculates the force adjustment amount by collecting ΔX and ΔV in real time. For example, if muscle relaxation causes the massage head 1 to shift by 2mm, it is converted into corresponding force compensation through the K value to ensure that the force always acts on the acupoint and achieves the target effect, avoiding the force deviation caused by the dynamic changes of the human body in static control.

[0064] The dynamic stiffness coefficient K and damping coefficient B can be flexibly adjusted according to the type of acupoint and physiological characteristics. For example, the K value of a pain point acupoint is slightly lower and the B value is slightly higher, which not only avoids the sudden increase in force when the position shifts and causes pain, but also makes the force transition more gentle when the speed changes through damping buffer.

[0065] The presence of the damping coefficient B ensures that the intensity adjustment is gradual rather than abrupt when the speed of the massage head 1 changes. For example, when the massage head 1 switches from a moving state to a stationary pressing state, the speed change ΔV is relatively large. The B value will buffer the intensity adjustment, avoiding the stinging or soreness caused by a sudden increase in intensity, making the massage process more in line with the body's tolerance rhythm.

[0066] The force compensation caused by positional deviation is accurately converted through the K value to avoid over-adjustment or under-adjustment. For example, a slight positional deviation (within 0.5mm) will only be compensated with a small amount of force, so that the force will not fluctuate greatly due to a small deviation, thus maintaining the stability of the massage experience.

[0067] In this embodiment, ΔX and ΔV can be directly acquired by the robot's position and speed sensors. K and B are adjustable parameters, facilitating optimization and calibration based on actual test data without requiring complex calculation logic, thus reducing the engineering implementation difficulty of the control system. The parameters are flexibly adjustable, allowing for the preset of corresponding K and B values ​​for different populations and acupoint types. For example, the K value can be generally lowered and the B value generally increased for children, further broadening the applicability of the technical solution and improving its feasibility.

[0068] More specifically, in one embodiment of the present invention, the dynamic stiffness coefficient K The adaptive adjustment method is as follows: When the actual contact force is less than the maximum permissible contact force

[0069] When the actual contact force is greater than or equal to the maximum permissible contact force ; in, For the maximum permissible contact force, For actual contact force, This is the initial stiffness coefficient.

[0070] In this embodiment, the maximum permissible contact force is used as the key threshold. This threshold is derived from the target force parameter calculated by the force control model. For example, the maximum permissible contact force for core acupoints and users with thick muscles is higher than that for users with peripheral acupoints and poor skin elasticity, ensuring that the threshold itself fits the characteristics of the individual and acupoint.

[0071] The triggering conditions for the dynamic stiffness coefficient K fall into two categories: one is when the actual contact force is less than the maximum allowable contact force (safe and effective range), and the other is when the actual contact force is greater than or equal to the maximum allowable contact force (critical / super-safe range). These two categories correspond to different adjustment directions: When the actual contact force is less than the maximum permissible contact force: This is within a safe range but not yet at the optimal stimulation level, and the adjustment method tends to enhance the force adaptation. By increasing the dynamic stiffness coefficient K, the efficiency of converting the positional offset ΔX of the massage head 1 into the force adjustment ΔF is improved. For example, positional offset caused by slight muscle deformation can be quickly compensated for by a higher K value, ensuring that the stimulation depth meets the standard and avoiding ineffective treatment due to insufficient contact force.

[0072] When the actual contact force is greater than or equal to the maximum permissible contact force: at this point, the safety tolerance limit has been approached or exceeded, and the adjustment method shifts to safety buffering. By reducing the dynamic stiffness coefficient K, the correlation between positional offset and force adjustment is weakened. Even if the massage head 1 has a slight offset, the force adjustment amount ΔF will not increase significantly, preventing the contact force from exceeding the limit further. At the same time, it allows time for the coordinated adjustment of temperature and force in the closed-loop feedback.

[0073] The initial stiffness coefficient serves as the adjustment benchmark, and its value can be preset based on the type of acupoint and physiological characteristics (e.g., the initial K value of core acupoints is higher than that of pain point acupoints) to ensure that the starting point of adaptive adjustment matches the basic scenario requirements.

[0074] More specifically, in one embodiment of the present invention, contact force data and moxibustion temperature data are collected in real time. These two types of data are direct feedback indicators for force control and temperature control. Contact force data reflects whether the massage effect meets the standard, and moxibustion temperature data reflects whether the warm stimulation is within a safe and comfortable range. Simultaneous collection of both ensures the synchronization of the basic data for coordinated control. The closed-loop feedback mechanism includes: Real-time acquisition of contact force data and moxibustion temperature data; calculation of the deviation between actual and target values. The target value comes from the target massage force output by the force control model and the moxibustion temperature based on collaborative matching. The deviation value is used to quantify the gap between the actual effect and the expected effect, providing a clear basis for subsequent adjustments.

[0075] When the deviation exceeds the preset threshold, the force of the massage head 1 and the heating power of the moxibustion component are adjusted to achieve dual parameter regulation of force and temperature. For example, when the temperature is too high, in addition to reducing the heating power, the massage force can also be appropriately reduced to avoid the dual stimulation of high temperature and high force; when the contact force is insufficient, in addition to increasing the massage force, the temperature can also be slightly increased to assist muscle relaxation, making the adjustment more closely coordinated and controlled.

[0076] To match the real-time nature of human physiological feedback, feedback adjustment is completed every 50-100ms to ensure that the force deviation does not exceed ±0.5N and the temperature deviation does not exceed ±0.3℃. Since the human body responds quickly to changes in force and temperature, high-frequency adjustment can correct deviations before they are perceived by the body. For example, if a sudden increase in contact force is caused by muscle tension, the force can be reduced within 100ms to avoid the accumulation of discomfort.

[0077] All adjustments are aimed at not exceeding force and temperature deviations, ensuring that regardless of changes in the human body's condition, the final output force and temperature can remain stable within a very small fluctuation range near the target values.

[0078] Therefore, by precisely controlling the deviation, ineffective operations are avoided. For example, the intensity is always within the effective stimulation range, and the temperature is always within the safe and comfortable range. The treatment will not be ineffective due to deviation, nor will it exceed the safety threshold due to deviation.

[0079] More specifically, in one embodiment of the present invention, the acupoint massage intensity control method further includes a collision detection step: monitoring the sudden change in force on the massage head 1 through a torque observer; when the torque change exceeds a preset sudden change threshold, controlling the massage head 1 to immediately reduce the force and adjust its position to avoid injury to the human body.

[0080] Specifically, a torque observer is used as the core monitoring component to capture transiently changing signals with high sensitivity. Compared to force sensor 2, which focuses on the continuous acquisition of conventional contact forces, the torque observer can quickly identify sudden changes in force within a short period of time, such as when a person suddenly turns over, when the massage head 1 accidentally touches hard tissues such as bones or joints, or when a sudden jamming of mechanical transmission causes a sudden change in force. It can accurately distinguish between the gradual force caused by the natural deformation of muscles and the sudden force caused by collisions, thus avoiding misjudgment.

[0081] The risk trigger condition is set when the change in torque exceeds a preset threshold. The threshold setting must take into account human safety tolerance and mechanical characteristics. For example, the preset torque mutation threshold is 5N. m, when the torque was monitored to change from 8N ​​within 10ms. m surged to 14N When the value is m, it is immediately identified as a collision risk to ensure the clarity and consistency of risk identification.

[0082] During use, the massage head 1 is first controlled to instantly reduce the force, quickly terminating the continuous impact of sudden force on the body; simultaneously, the position of the massage head 1 is adjusted to move away from the current risk contact point, thus doubly avoiding injury. After the emergency action is executed, the system does not shut down directly, but instead recalibrates the acupoint position through image recognition and recalculates the safe force based on physiological characteristic data. The closed-loop feedback mechanism gradually restores normal conditioning, ensuring safety while minimizing the impact on the conditioning process.

[0083] In this way, by accurately distinguishing between sudden and gradual force application, false triggering is avoided. The torque observer's precise identification of "sudden change signals" can eliminate interference from gradual force application such as natural muscle contraction and breathing fluctuations during massage, ensuring that emergency response is only activated when there is a genuine risk of collision, without affecting the continuity and comfort of regular treatment.

[0084] More specifically, in one embodiment of the present invention, the massage head 1 includes an EMS massage head, wherein the pulse current intensity of the EMS massage head is controlled in conjunction with the massage force parameter, and the greater the force, the higher the corresponding pulse current intensity level, and the pulse current intensity is adjustable in 3-5 levels.

[0085] Specifically, the greater the intensity of the massage, the higher the intensity of the pulsed current, which aligns with the physiological response of the human body. When the intensity of a physical massage is increased, the muscles are in a state of tension or stimulation. Simultaneously enhancing the electrophysiological stimulation of the pulsed current allows the stimulation to penetrate deeper into muscle fibers and nerve endings, avoiding the problem of insufficient surface stimulation from a single physical massage.

[0086] By setting 3-5 adjustable pulse current intensities, the range avoids the complexity of continuous current intensity adjustment and covers the tolerance of different users. For example, users with sensitive constitutions can choose 1-2 low current levels, while users with high tolerance can choose 3-5 high current levels, which is compatible with the physiological characteristic data mentioned above (such as acupoint sensitivity coefficient).

[0087] More specifically, in one embodiment of the present invention, the acupoint massage intensity control method further includes a physiotherapy mode switching step: automatically switching between acupoint opening mode, strengthening mode and acupoint sealing mode based on the user's physiotherapy stage. The target massage intensity, moxibustion temperature and feedback adjustment frequency are different in different modes. The acupoint opening mode has a gentle intensity, the strengthening mode has a strong intensity, and the acupoint sealing mode has a gentle intensity.

[0088] This technical solution deeply integrates traditional Chinese medicine conditioning logic with intelligent control through a phased therapy mode switching of acupoint opening, strengthening, and sealing. It achieves an upgrade from single-parameter control to dynamic adaptation throughout the entire process, with the technical process conforming to the body's physiological rhythms. The opening mode, as the initial stage, aims to awaken acupoints and relax tissues. Therefore, it is designed with gentle intensity to avoid muscle resistance caused by strong initial stimulation. The strengthening mode is the core stage of conditioning, which aims to deeply stimulate acupoints and exert conditioning effects. It requires strong intensity to ensure the penetration of stimulation. The closing mode, as the final stage, focuses on consolidating the effect and gently ending the treatment. The intensity tends to be gentle to avoid over-sensitivity of acupoints and reduce subsequent discomfort.

[0089] Preferably, in this embodiment, the acupoint opening mode needs to be gentle and adaptable, with low intensity set at 8-15N, medium-low temperature at 38-40℃, and low-frequency feedback at 100ms; the enhancement mode needs to be precise and effective, with high intensity set at 20-30N, medium-high temperature at 41-43℃, and high-frequency feedback at 50ms; and the acupoint sealing mode needs to be smooth and gradual, with medium-low intensity set at 10-18N, medium temperature at 39-41℃, and medium-frequency feedback at 80ms.

[0090] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of this disclosure, and all such changes and modifications will fall within the scope of protection of this invention.

Claims

1. A method for controlling the intensity of acupoint massage in an intelligent moxibustion robot, characterized in that, Includes the following steps: S1: Collect human images and physiological feature data of the target area, locate the acupoints and their types through image recognition algorithms, and obtain acupoint location information; S2: Based on the acupoint positioning information, control the massage head of the intelligent moxibustion robot to move to the target acupoint, and collect the contact force data between the massage head and the human body in real time through the force sensor; S3: Obtain the real-time working temperature data of the moxibustion component, and construct a force control model by combining the acupoint positioning information, physiological characteristic data and contact force data; S4: Based on the force control model, the target massage force parameters are calculated using an impedance control algorithm; S5: Control the massage head to output corresponding force according to the target massage force parameter, and adjust the massage force and moxibustion temperature in real time through a closed-loop feedback mechanism to achieve coordinated control.

2. The method for controlling the acupoint massage intensity of an intelligent moxibustion robot according to claim 1, characterized in that, The physiological characteristic data includes body weight, muscle thickness, skin elasticity, and acupoint sensitivity coefficient, which is obtained through a preset exploratory massage process.

3. The method for controlling the acupoint massage intensity of an intelligent moxibustion robot according to claim 1, characterized in that, The acupoint location information includes acupoint coordinates, the meridian to which the acupoint belongs, and the acupoint type. The acupoint type includes core acupoints, peripheral acupoints, and pain point acupoints.

4. The method for controlling the acupoint massage intensity of an intelligent moxibustion robot according to claim 1, characterized in that, In step S3, the process of constructing the force control model include: Step S 31 The basic pressure range is determined based on the type of acupoint, with the basic pressure range for core acupoints being higher than that for peripheral acupoints. Step S 32 The basic force range is corrected by combining physiological characteristic data; the greater the muscle thickness, the greater the force correction coefficient. Step S 33 Introducing the temperature-influencing factor of moxibustion: When the moxibustion temperature is in the range of 38-43℃, the intensity adjustment coefficient increases by 5%-8% for every 1℃ increase in temperature.

5. The method for controlling the acupoint massage intensity of an intelligent moxibustion robot according to claim 1, characterized in that, The impedance control algorithm uses the following formula: ΔF = K·ΔX + B·ΔV Where ΔF is the force adjustment amount, K is the dynamic stiffness coefficient, ΔX is the massage head position offset amount, B is the damping coefficient, and ΔV is the change in the massage head movement speed.

6. The method for controlling the intensity of acupoint massage in an intelligent moxibustion robot according to claim 5, characterized in that, The dynamic stiffness coefficient K The adaptive adjustment method is as follows: When the actual contact force is less than the maximum permissible contact force When the actual contact force is greater than or equal to the maximum permissible contact force ; in, For the maximum permissible contact force, For actual contact force, This is the initial stiffness coefficient.

7. The method for controlling the acupoint massage intensity of an intelligent moxibustion robot according to claim 1, characterized in that, The closed-loop feedback mechanism includes: Real-time acquisition of contact force data and moxibustion temperature data; calculation of the deviation between actual and target values. When the deviation exceeds the preset threshold, adjust the force of the massage head and the heating power of the moxibustion component; Feedback adjustment is completed every 50-100ms to ensure that the force deviation does not exceed ±0.5N and the temperature deviation does not exceed ±0.3℃.

8. The method for controlling the intensity of acupoint massage in an intelligent moxibustion robot according to claim 1, characterized in that, It also includes a collision detection step: a torque observer monitors sudden changes in the force on the massage head. When the detected torque change exceeds a preset threshold, the massage head is controlled to immediately reduce the force and adjust its position to avoid injury to the human body.

9. The method for controlling the intensity of acupoint massage in an intelligent moxibustion robot according to claim 1, characterized in that, The massage head includes an EMS massage head, and the pulse current intensity of the EMS massage head is controlled in conjunction with the massage force parameter. The greater the force, the higher the corresponding pulse current intensity level, and the pulse current intensity is adjustable in 3-5 levels.

10. The method for controlling the acupoint massage intensity of an intelligent moxibustion robot according to claim 1, characterized in that, It also includes a physiotherapy mode switching step: automatically switching between acupoint opening mode, strengthening mode and acupoint sealing mode based on the user's physiotherapy stage. The target massage intensity, moxibustion temperature and feedback adjustment frequency are different in different modes. The acupoint opening mode is gentle, the strengthening mode is strong, and the acupoint sealing mode tends to be gentle.

Images