A 5D movement self-learning massage method and massage chair
Through the 5D movement self-learning method, the massage velocity curve is automatically adjusted using sensors and processing units, which solves the problem that the existing massage movement cannot match the user's strength preferences, and achieves the precise matching and personalized experience of massage velocity.
Patent Information
- Application Number
- CN202410910507.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-07-09
AI Technical Summary
The existing massage movement cannot automatically adjust the massage force to match the user's strength preferences. The user needs to manually adjust or rely on uncertain automatic scanning, resulting in poor massage force and insufficient user experience.
Using the 5D movement self-learning method, the pressure and position data of the massage head are detected by force sensors and position sensors, and the massage force curve is generated and adjusted through the central processing unit. Combined with genetic algorithms and nonlinear regression, it can independently learn and match the user's massage. Strength preference.
It achieves accurate matching of massage strength, improves the comfort and personalization of massage, ensures that the massage strength is stable within a specific range, and reduces user discomfort and the control complexity of massage movements.
Smart Images

Figure CN118453367B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of massage chairs, in particular to a 5D movement self-learning massage method and a massage chair. Background Art
[0002] The main massage techniques include pressing, rubbing, pushing, holding, kneading, and pinching. Among them, "kneading" is to use fingers or palms to make rotational movements on the skin or acupuncture points, and "pinching" is to use the thumb and other fingers to symmetrically squeeze the treated area. The massage movement usually has two massage heads. When they "knead" synchronously in clockwise and counterclockwise directions, "pinching" is achieved in the process of relative squeezing (moving towards each other) and stretching (moving in the opposite direction). Such an overall movement of the movement is combined into a set of basic techniques of the massage movement - kneading.
[0003] During the kneading process of the two massage heads of the massage movement, the time series of the interaction force between the massage heads and the body (i.e., massage intensity) will show a periodic regularity of alternating significant peaks (when squeezed to the closest) and valleys (when stretched to the farthest). The periodicity is related to the rotation of the massage heads, and the specific size of the regular curve and its peak and valley values are also affected by the massage part, the posture of the massage chair, the stroke of the massage head ejection guide surface, etc. In a certain part and posture, if you want to change the size of the massage intensity, the usual operation is to adjust the ejection stroke of the massage head, usually manually by the user or automatically controlled by the massage movement.
[0004] At present, for massage mechanisms that do not have the ability to sense and adjust massage force, users are required to manually adjust the massage gears in a specific massage chair posture. Each gear corresponds to a stroke of the massage head ejecting the guide surface to adapt to the user's own strength preference for a specific massage technique. Compared with the maximum ejection stroke of the massage head, generally only a small number of gears are set, and the user's options are limited; and if the span between the gears is large, it is not easy to adjust to the best gear.
[0005] As for the automatic control of the massage movement, there is a method to scan the back massage intensity, obtain the tolerance threshold, and then adapt it to the corresponding massage technique. However, there are many uncertain factors in a single scan, and there may be a large deviation in the tolerance threshold; and the interaction force under different massage techniques also has different change characteristics and regular curves, so it is not appropriate to generalize. Summary of the invention
[0006] The main purpose of the present invention is to overcome the defect in the prior art that the massage intensity of the kneading technique of the massage movement cannot automatically match the user's intensity preference, and to propose a 5D movement self-learning massage method and massage chair, which can intelligently obtain the limit value correspondence matching the user's massage intensity preference through autonomous learning, which is more personalized.
[0007] The present invention adopts the following technical solution:
[0008] A 5D movement self-learning massage method, comprising:
[0009] The massage movement drives the massage head on the massage swing arm to apply a roughly circular rotation massage to a part of the user's body, with each circular rotation massage as a cycle, and the force sensor detects the pressure of the user's body on the massage head as the massage force;
[0010] The massage head is moved to other parts of the user's body, and the massage is continued to be rotated in a roughly circular trajectory, with each circular rotation massage as a cycle, and the force sensor is used to detect the pressure exerted by the user's body on the massage head;
[0011] The position sensor is used to detect the spatial position, i.e. the rotational position, of the massage head during the above massage action in real time;
[0012] The massage force data and the rotation position data detected by the force sensor and the position sensor are transmitted to the central processing unit, and the central processing unit is used to process and generate massage force curves of different parts of the user's body in each rotation massage cycle;
[0013] The central processing unit controls the massage movement to push out or withdraw the massage head in the direction toward the user's body, so as to adjust the massage intensity applied by the massage head to the body parts in different rotation massage cycles.
[0014] Also includes:
[0015] Identify a specific user;
[0016] When the massage movement is initially used, a massage technique is performed for a specific user whose identity is identified based on a massage intensity curve in a rotation massage cycle corresponding to different human body parts, and an upper and lower limit curve reflecting the preference of different human body parts of the specific user is obtained through autonomous learning based on the massage intensity data and rotation position data collected in real time;
[0017] When the massage movement is subsequently used, the central processing unit controls the massage head to perform massage techniques within a range enclosed by the upper and lower limit curves corresponding to the preference for different parts of the human body.
[0018] The central processing unit uses nonlinear regression to establish a corresponding relationship curve between the massage force and rotation position of the massage head based on the data detected by the force sensor and the position sensor, and obtains the massage force curves of different parts of the human body in each rotation massage cycle after adjusting the parameters.
[0019] The massage force curve includes an initial upper limit value curve and an initial lower limit value curve:
[0020] f1=a1*h1(b1*(s-c1))+d1;
[0021] f2=a2*h2(b2*(s-c2))+d2;
[0022] Among them, f1 and f2 are the upper limit and lower limit of the massage intensity respectively, h1(.) and h2(.) represent the kernel functions of the initial upper limit curve and the initial lower limit curve respectively, a1 is the height of the initial upper limit curve, a2 is the height of the initial lower limit curve, b1 is the steepness of the initial upper limit curve, b2 is the steepness of the initial lower limit curve, c1 is the offset of the initial upper limit curve, c2 is the offset of the initial lower limit curve, d1 is the translation of the initial upper limit curve, and d2 is the translation of the initial lower limit curve.
[0023] The different parts of the human body include shoulders, waist and buttocks, which correspondingly include the massage intensity curve and the preferred upper and lower limit curves of the shoulders in each rotational massage cycle, the massage intensity curve and the preferred upper and lower limit curves of the waist in each rotational massage cycle, and the massage intensity curve and the preferred upper and lower limit curves of the buttocks in each rotational massage cycle.
[0024] The specific calculation method of the upper and lower limit curves reflecting the preferences of different body parts of the specific user is as follows:
[0025] Divide the collected rotation position data into several intervals, and obtain the maximum and minimum values of the massage intensity data in each interval;
[0026] Calculating the sum of distances between the maximum values in a plurality of the intervals and the corresponding upper limit values of the massage intensity curve, and the sum of distances between the minimum values in a plurality of the intervals and the corresponding lower limit values of the massage intensity curve;
[0027] A genetic algorithm is used for feedback learning and autonomous adjustment to obtain a set of control parameters of the massage intensity curve so as to minimize the sum of the distances, and the massage intensity curve with the control parameters is used as the upper and lower limit curves of the preference.
[0028] When the central processing unit determines that the real-time massage intensity exceeds the upper and lower limit curve range of the preference, the central processing unit controls the massage head on the massage movement to push out or withdraw so that the real-time massage intensity is within the upper and lower limit curve range of the preference; or, when a specific user actively increases pressure on the massage head, and the central processing unit determines that the real-time massage intensity exceeds the upper and lower limit curve range of the preference, the central processing unit controls the massage head on the massage movement to push out or withdraw so that the real-time massage intensity is within the upper and lower limit curve range of the preference.
[0029] It also includes adjusting the strength of the massage head according to the expectations of a specific user, specifically including:
[0030] The massage movement analyzes the adjustment direction, frequency and rate of change of the massage intensity before adjustment of the massage head's ejection stroke to continuously detect whether the user has an avoiding action or a pushing action. If there is an avoiding action or a pushing action, it means that the user expects to adjust the massage intensity of the massage head.
[0031] The avoiding action refers to the behavior of the user lifting the massaged body part relative to the massage movement to reduce or even avoid contact with the massage head, and the avoiding action is regarded as the user's expectation to obtain a relatively smaller massage intensity; the pushing action refers to the behavior of the user pressing back on the massaged body part to push back the massage head in an antagonistic manner to increase contact with the massage head, and the pushing action is regarded as the user's expectation to obtain a relatively greater massage intensity.
[0032] If the user's desire to adjust the massage intensity of the massage head is still detected after the intensity of the massage head is adjusted according to the user's desire, the control parameters of the preferred upper and lower limit curves are controlled to match the massage intensity preferred by the user.
[0033] A massage chair comprises a massage movement and a central processing unit, and adopts the 5D movement self-learning massage method.
[0034] The device also includes a display module, which displays in real time the massage intensity applied by the massage head to the body parts during different rotation massage cycles and displays the upper and lower limit curves of the preference.
[0035] It also includes an adjustment module, which is used to adjust the real-time massage force and the speed of the massage head movement, thereby changing the shape of the massage force curve; the adjustment module also allows the user to manually adjust the control parameters of the preference upper and lower limit curves.
[0036] It can be seen from the above description of the present invention that, compared with the prior art, the present invention has the following beneficial effects:
[0037] In the present invention, a massage movement is used to drive the massage head on the massage swing arm to apply a roughly circular rotation massage to different parts of the user's body. With each circular rotation massage as a cycle, a force sensor is used to detect the pressure exerted on the massage head by the user's body, and a position sensor is used to detect the rotation position of the massage head in the above massage action in real time; the detected pressure data and rotation position data are transmitted to the central processing unit, which is processed to generate a massage force curve for different parts of the user's body in each rotation massage cycle; the massage force curve is used as an initial upper and lower limit curve for subsequent autonomous learning.
[0038] Among them, the basic massage trajectory is set as a roughly circular rotation trajectory in order to imitate the massage technique of "kneading" the human body by the massage head. And "kneading" is the most common massage technique in the entire massage procedure. Due to the large differences in the curves of the human back and buttocks, the pressure changes on the massage head vary greatly during the "kneading" technique cycle of different parts. However, for a specific body part, the "kneading" technique shows obvious regular changes in peaks and valleys. This regular change is caused by the swing arm of the massage movement driving the massage head to move. As long as the peak and valley values of the total pressure in the "kneading" technique cycle are controlled well, the massage intensity of the "kneading" technique cycle can be effectively adjusted. As long as the intensity of each "kneading" technique cycle can be well controlled, the comfort of the overall massage technique will be greatly improved.
[0039] In the present invention, the massage intensity curves of different human body parts in each rotational massage cycle are processed according to the collected rotational position data and massage intensity data to obtain the upper and lower limit curves preferred by different human body parts of a specific user. The massage technique is performed using the preferred upper and lower limit curves, and the limit correspondence matching the user's massage intensity preference is obtained intelligently through autonomous learning, so that consumers have a better and more personalized user experience.
[0040] In the present invention, before the massage movement leaves the factory, the rotation position data and massage strength data of the massage head when performing kneading techniques on different body parts of several volunteers are collected on a large scale, and the corresponding relationship curve between the massage strength and the rotation position of the massage head is established by using nonlinear regression to obtain the massage strength curves of different body parts of the user in each rotation massage cycle, and the initial upper and lower limit curves are obtained after parameter adjustment. The corresponding relationship obtained based on the large-scale data can better meet the changing trend and characteristics of the massage strength during kneading, and can also ensure that the kneading technique after control will not be deformed.
[0041] In the present invention, the upper and lower limit curves mean that when the massage head is in each rotational position, there is an upper limit value of the massage intensity and a lower limit value of the massage intensity corresponding to the rotational position, and different parts of the human body are correspondingly provided with massage intensity curves and preferred upper and lower limit curves, thereby realizing full-process adjustment of the massage intensity of different parts, increasing the continuity and smoothness of the massage movement in control, and ensuring that consumers will not experience discomfort in use.
[0042] In the present invention, the collected data is used to perform feedback learning and autonomous adjustment on the massage force curve using a genetic algorithm to obtain a set of control parameters so that the collected data is within the corresponding relationship range, thereby obtaining a preferred upper and lower limit curve.
[0043] When the massage movement of the present invention is used subsequently, massage techniques are performed on different parts of the human body based on the corresponding preferred upper and lower limit curves, and when it is sensed during the whole process that the actual massage intensity is not within the range of the preferred upper and lower limit curves, the massage movement can automatically adjust to an appropriate massage head top stroke; if after adjusting the intensity of the massage head according to the user's expectations, the user's desire to adjust the massage intensity is still detected, the control parameters of the preferred upper and lower limit curves are controlled to match the massage intensity preferred by the user, thereby more accurately meeting the user's needs.
[0044] The massage chair of the present invention adopts human-computer interaction during the entire use process, and can display in real time the massage intensity applied by the massage head to body parts during different rotation massage cycles and the preferred upper and lower limit curves, etc., and visually show the user the specific change trend of the corresponding relationship between the upper and lower limits along with the control parameters such as height, steepness and translation, etc., and can also appropriately fine-tune the corresponding relationship between the upper and lower limits to match their massage intensity preferences and wishes, which is more humane. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 This is a schematic diagram of the rotation trajectories of the two massage heads of the present invention;
[0046] Figure 2 is a flow chart of the method of the present invention;
[0047] Figure 3 The corresponding relationship between massage intensity and rotation position and the initial upper and lower limit curve diagram;
[0048] Figure 4 This is a comparison chart of the height of the upper limit curve before and after adjustment;
[0049] Figure 5 This is a comparison chart of the upper limit value curve before and after the kurtosis adjustment;
[0050] Figure 6 This is a comparison chart before and after the offset adjustment of the upper limit value curve;
[0051] Figure 7 This is a comparison chart of the upper limit value curve before and after the translation adjustment;
[0052] Figure 8 This is a comparison chart of the upper limit curve before and after adjustment of the translation and height;
[0053] Fig. 9 It is a relationship diagram between the massage intensity curve and the upper and lower limit curves of preference;
[0054] Fig.10 It is a schematic diagram of the display module and the adjustment module.
[0055] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. DETAILED DESCRIPTION
[0056] The present invention is further described below through specific implementation modes.
[0057] The massage swing arm of the massage movement of the present invention is connected with two massage heads, and the two massage heads cooperate to realize massage techniques such as pressing, rubbing, pushing, holding, kneading, and pinching. When the two massage heads perform "kneading" synchronously in clockwise and counterclockwise directions respectively, "pinching" is realized in the process of relative squeezing, i.e., moving toward each other, and stretching, i.e., moving in the opposite direction. Such an overall action of the massage movement is combined into a set of basic techniques of the massage movement - kneading.
[0058] The rotational position of the two massage heads in the present invention refers to the position of the massage heads on the circular trajectory during the rotational motion of the kneading technique. Figure 1 , the rotation position of the two massage heads when they are about to move from the farthest relative distance to the closest relative distance is -β (β>0), the rotation position at the closest relative distance is 0, and the rotation position when they are about to reach the farthest relative distance is β. In this way, the rotation position of the massage head will reciprocate between -β and β for each rotation of the massage head.
[0059] The massage intensity in the present invention refers to the interaction force between the massage head and the user's body sensed by the massage core through the sensor.
[0060] See also Figure 2 The present invention proposes a 5D movement self-learning massage method, which mainly includes the following parts: how to obtain the massage force curves of different human body parts in each rotation massage cycle, how to obtain the upper and lower limit curves reflecting the preferences of different human body parts of a specific user, and how to adjust the massage technique according to the upper and lower limit curves.
[0061] The massage force curves of different parts of the human body in each rotation massage cycle are obtained by the following method:
[0062] The massage movement drives the massage head on the massage swing arm to apply a roughly circular rotation massage to one part of the user's body. Each circular rotation massage is a cycle, and the force sensor is used to detect the pressure exerted by the user's body on the massage head. The massage head is moved to other parts of the user's body, and the massage is continued to be applied with a roughly circular rotation massage. Each circular rotation massage is a cycle, and the force sensor is used to detect the pressure exerted by the user's body on the massage head, which is the massage force. The position sensor is used to detect the spatial position of the massage head in the above massage action in real time.
[0063] The pressure data and spatial position data detected by the force sensor and the position sensor are transmitted to the central processing unit, and the central processing unit is used to process and generate massage force curves for different parts of the user's body in each rotation massage cycle. The central processing unit controls the massage movement to push out or withdraw the massage head in the direction toward the user's body to adjust the massage force applied by the massage head to the body parts in different rotation massage cycles.
[0064] In the present invention, the massage intensity curve is a curve of the relationship between the massage intensity of the massage head and the rotational position, which means that when the massage head performs massage techniques, when the massage head is in each rotational position, there is an initial upper limit value of the massage intensity and an initial lower limit value of the massage intensity corresponding to the rotational position.
[0065] In actual application, the specific method for obtaining the massage intensity curve is:
[0066] Before the massage movement leaves the factory, the rotational position data and massage intensity data of the massage head when performing kneading techniques on different body parts of several volunteers are collected. Nonlinear regression is used to establish a corresponding relationship curve between the massage intensity and rotational position of the massage head. After parameter adjustment, the massage intensity curves of different body parts in each rotation massage cycle are obtained.
[0067] Among them, a large number of volunteers are used, and the volunteers are not distinguished by age and body shape. Different human body parts include shoulders, waist, hips and other back parts. Nonlinear regression accurately depicts the overall change law of massage intensity with respect to the rotation position of the massage head based on data such as the rotation position of the massage head and the massage intensity. Further, it can be a support vector regression based on a specific kernel function. The specific kernel function uses the radial basis function form h(x)=exp(-λ*x^2), where λ>0.
[0068] In this step, parameter adjustment refers to obtaining an initial upper limit value curve and an initial lower limit value curve of the massage intensity of the massage head with respect to the rotation position by adjusting relevant parameters of the corresponding relationship without changing the changing law of the corresponding relationship between the rotation position of the massage head and the massage intensity, and making most of the collected data just fall between the initial upper limit value curve and the initial lower limit value curve.
[0069] Specifically, based on the selected specific kernel function, for example, for the data collected at the waist, it is assumed that the corresponding relationship under support vector regression is:
[0070] f=α*h(s)+γ=α*exp(-λ*s^2)+γ;
[0071] Among them, f represents the massage intensity, s represents the rotation position of the massage head, h(.) represents the kernel function, α>0, λ>0, γ>0, and parameter adjustment is to determine the specific values of α, λ, γ>0.
[0072] Then, after adjusting the parameters, the corresponding relationship between the upper and lower limits is expressed as:
[0073] f1=h1(s)=α1*exp(-λ1*s^2)+γ1;
[0074] f2=h2(s)=α2*exp(-λ2*s^2)+γ2;
[0075] Further expansion to:
[0076] f1=a1*h1(b1*(s-c1))+d1;
[0077] f2=a2*h2(b2*(s-c2))+d2;
[0078] Among them, f1 and f2 are the upper limit and lower limit of massage intensity respectively, h1(.) and h2(.) represent the kernel functions of the initial upper limit curve and the initial lower limit curve respectively; a1 is the height of the initial upper limit curve, a2 is the height of the initial lower limit curve, a1 and a2 (both values are greater than 0) are parameters for controlling the height of the curve shape respectively; b1 is the steepness of the initial upper limit curve, b2 is the steepness of the initial lower limit curve, b1 and b2 (both values are greater than 0) are parameters for controlling the steepness of the curve shape respectively; c1 is the offset of the initial upper limit curve, c2 is the offset of the initial lower limit curve, c1 and c2 (values are always equal) are the offset parameters corresponding to s when the control curve reaches the peak synchronously; d1 is the translation of the initial upper limit curve, d2 is the translation of the initial lower limit curve, d1 and d2 (both values are greater than 0) are the translation parameters corresponding to the massage intensity of the control curve respectively.
[0079] The process of adjusting α, λ, γ to obtain α1, λ1, γ1 and α2, λ2, γ2 can be done manually. For example, when set to the initial state, the height values a1 and a2, the steepness values b1 and b2 are all 1, the offset values c1 and c2 are the rotation positions of the two massage heads when the relative distance is the shortest (as above, they are all 0), and the translation values d1 and d2 are also 0.
[0080] The process of adjusting α, λ, γ to obtain α1, λ1, γ1 and α2, λ2, γ2 can also adopt the autonomous learning method, that is, by adjusting the corresponding relationship of α1, λ1, γ1 and α2, λ2, γ2, most of the collected data are just between the initial upper limit curve and the initial lower limit curve. Figure 3 , which includes the correspondence between the massage strength and the rotation position of the massage head, as well as the initial upper limit value curve and the initial lower limit value curve.
[0081] Similarly, for other back parts such as shoulders and buttocks, corresponding upper and lower limit correspondences and control parameters such as height, steepness, offset, translation and their values can also be obtained, including the massage intensity curve of the shoulders, the massage intensity curve of the waist, the massage intensity curve of the buttocks, etc.
[0082] In this step, the upper and lower limit value correspondences obtained above that match the massage intensity preferences of the volunteer group for various parts of the back are equipped in all factory-set massage movements as their initial upper and lower limit value correspondences.
[0083] How to obtain the upper and lower limit curves reflecting the preferences of different human body parts of a specific user includes the following:
[0084] Identify the identity of a specific user. When the massage movement is initially used, massage techniques are performed on the specific user based on the massage force curve of different body parts in each rotation massage cycle. Based on the pressure data and rotation position data collected in real time, autonomous learning is performed to obtain the upper and lower limit curves preferred by different body parts of the specific user.
[0085] In this step, the specific user is a specific single user or a small number of users who perform a shorter-term experiential massage. During the massage movement executing the massage technique, when the massage head is in a certain rotation position, when the perceived massage force is greater than the upper limit value of the massage force curve, the massage head top stroke is appropriately reduced; when the perceived massage force is less than the lower limit value of the massage force curve, the massage head top stroke is appropriately increased.
[0086] Autonomous learning refers to the continuous and autonomous adjustment of control parameters such as height, steepness, offset, and translation on the basis of feedback learning (the corresponding relationship between the upper and lower limits will also change accordingly), until most of the data during the experiential massage is just between the corresponding upper and lower limits.
[0087] See also Fig. 9 , the specific calculation method of the upper and lower limit curves of different human body parts preferences of a specific user is as follows:
[0088] a1 divides the collected rotation position data into several intervals (intervals are separated by vertical lines in the figure), and obtains the maximum and minimum values of the massage intensity data in each interval (the square circle at the top of the figure is the maximum value, and the circle at the bottom is the minimum value). The value range of the rotation position of the massage head (-β to β) can be divided into several intervals, and the number of intervals can be set according to actual conditions and is not limited here.
[0089] a2 calculates the sum of the distances between the maximum values in several intervals and the upper limit value of the corresponding massage intensity curve, and the sum of the distances between the minimum values in several intervals and the lower limit value of the corresponding massage intensity curve.
[0090] Specifically, the sum of the distances between the maximum value in the interval and the upper limit of the corresponding massage intensity curve is calculated by first calculating the distance between the maximum value in each interval and the upper limit corresponding to the initial upper limit curve, and then adding the distances of each interval. Similarly, the sum of the distances between the minimum values in several intervals and the lower limit of the corresponding massage intensity curve is calculated by first calculating the distance between the minimum value in each interval and the lower limit corresponding to the initial lower limit curve, and then adding the distances of each interval.
[0091] a3 uses genetic algorithm for feedback learning and autonomous adjustment to obtain a set of control parameters of massage intensity curves so that the sum of distances is minimized, and the upper and lower limit curves with control parameters are used as preferred upper and lower limit curves.
[0092] The goal of this step is to find a set of control parameters so that the sum of the distances between the minimum value in several intervals and the lower limit value of the corresponding massage intensity curve is minimized, and the sum of the distances between the maximum value in several intervals and the upper limit value of the corresponding massage intensity curve is minimized. In the process of feedback learning and autonomous adjustment using intelligent optimization algorithms such as genetic algorithms, the optimal evolution and re-evolution are performed, and after a certain number of evolutions, a set of control parameters such as height, steepness, offset, and translation that are closest to achieving the goal are obtained.
[0093] The corresponding relationship between the upper and lower limits under the obtained control parameters, that is, the corresponding relationship between the upper and lower limits that matches the massage intensity preference of a specific user, can be used as the upper and lower limit curve of preference.
[0094] In the present invention, for different parts of the human body, preferred upper and lower limit curves are included for the shoulder, preferred upper and lower limit curves for the waist, and preferred upper and lower limit curves for the buttocks. The preferred upper and lower limit curves further include a preferred upper limit curve and a preferred lower limit curve.
[0095] After obtaining the upper and lower limit curves preferred by a specific user, the massage movement can perform massage techniques based on the corresponding upper and lower limit curves preferred for different parts of the human body during subsequent use. Specifically, it includes the following:
[0096] When the central processing unit determines that the real-time massage intensity exceeds the preferred upper and lower limit curve range, the central processing unit controls the massage head on the massage movement to push out or withdraw so that the real-time massage intensity is within the preferred upper and lower limit curve range; or, when a specific user actively increases pressure on the massage head, and the central processing unit determines that the real-time massage intensity exceeds the preferred upper and lower limit curve range, the central processing unit controls the massage head on the massage movement to push out or withdraw so that the real-time massage intensity is within the preferred upper and lower limit curve range.
[0097] In this step, the massage intensity of the massage head can also be adjusted according to the user's expectations. Specifically, it includes:
[0098] The massage movement analyzes the adjustment direction, frequency and rate of change of the massage intensity before adjustment of the massage head's top stroke to continuously detect whether the user has an avoiding action or a pushing action. If there is an avoiding action or a pushing action, it means that the user expects to adjust the massage intensity of the massage head.
[0099] In the present invention, the avoidance action refers to the behavior of the user lifting the massaged human body part relative to the massage movement to reduce or even avoid contact with the massage head. The avoidance action is regarded as the user's desire to obtain a relatively smaller massage intensity, and the massage movement controls the massage head to reduce the massage intensity. The top action refers to the behavior of the user pressing the massaged human body part backward to push the massage head back in a confrontational manner to increase the contact with the massage head. The top action is regarded as the user's desire to obtain a relatively greater massage intensity, and the massage movement controls the massage head to increase the massage intensity.
[0100] Furthermore, if after adjusting the strength of the massage head according to the user's expectation, the user's desire to adjust the massage strength of the massage head is still detected, the control parameters of the upper and lower limit curves of the adjustment preference are controlled to match the massage strength preferred by the user.
[0101] Specifically, when the massage head top stroke is frequently increased in a short period of time, if the massage force is still perceived to be less than the lower limit of the preferred upper and lower limit curve, and the massage force change rate before adjustment is small (not without significant change), then the user's intentional avoidance action is detected. At this time, the massage movement can obtain the upper and lower limit correspondence that matches the massage force preference of the specific user by autonomously reducing the height control parameter (associated with the lower limit correspondence), that is, the new preferred upper and lower limit curve.
[0102] When the massage head's top stroke is frequently reduced in a short period of time, if the massage force is still perceived to be greater than the upper limit of the preferred upper and lower limit curves, and the massage force change rate is extremely large before adjustment, it is detected that the user has an intentional top-up action. At this time, the massage movement can obtain the upper and lower limit correspondence that matches the massage force preference of the specific user by autonomously increasing control parameters such as the height and translation amount (associated with the upper limit correspondence), that is, the new preferred upper and lower limit curves.
[0103] Based on this, the present invention also proposes a massage chair, including a massage movement, and also includes necessary components such as a backrest, an airbag assembly, a guide rail, a control system, etc. The control system is provided with a central processing unit.
[0104] The present invention also includes a display module and an adjustment module. The display module displays in real time the massage intensity applied by the massage head to the body parts in different rotation massage cycles and displays the upper and lower limit curves of preference. The display module is also used to display the control parameters of the upper and lower limit curves of control preference and other necessary information, realizing an explicit mode of human-computer interaction.
[0105] The adjustment module is used to adjust the real-time massage intensity and the speed of the massage head movement, thereby changing the shape of the massage intensity curve; the adjustment module also allows users to manually adjust the control parameters of the upper and lower limit curves or other massage related parameters.
[0106] Specifically, the display module can be integrated into the display of the massage chair, and a strength adjustment button can be set. The display displays in real time the massage strength applied by the massage head to the body parts during different rotation massage cycles, and the strength adjustment button is used to adjust the size of the real-time massage strength, so that the user can intuitively feel the graphics corresponding to different massage strengths, which is convenient for the user to adjust. Furthermore, the strength adjustment button can also adjust the speed of the massage head movement, thereby changing the shape of the massage strength curve. Furthermore, the adjustment module also includes a speed adjustment button, which is used to adjust the speed of the massage head movement, thereby changing the shape of the massage strength curve.
[0107] The cooperation between the display module and the adjustment module enables the user to intuitively feel the human-computer interaction process, allowing the user to more conveniently adjust the upper and lower limit curves of preference, and cooperate with the 5D movement self-learning massage method to effectively improve the user's massage experience.
[0108] During the whole process of using the massage chair of the present invention, the user can view in real time the graph drawn by the correspondence between the actual value of the massage intensity of the massaged part and its upper and lower limits with respect to the rotation position of the massage head within a certain massage time window. The specific change trend of the upper and lower limit correspondence along with the control parameters such as height, steepness, and translation is displayed to the user in a visual manner. Thus, the user can be guided in an explanatory manner to make appropriate fine adjustments to the upper and lower limit correspondence of the massaged part manually based on their actual experience, so as to match their massage intensity preference and intention.
[0109] Among them, moderate fine-tuning refers to adjusting the values of control parameters such as height, steepness, and translation within a smaller range. Specifically, when the user has a massage demand for less restricted massage intensity when pinching according to the real-time massage intensity graph, he can get the corresponding visual guidance - reducing the height will relatively reduce the massage intensity when pinching, and the height can be fine-tuned manually to obtain the upper and lower limit values that match his massage intensity preference and willingness.
[0110] When the user has a need for a higher massage intensity with less restriction in the middle section of kneading, squeezing and stretching based on the real-time massage intensity graph, he or she can get the corresponding visual guidance - reducing the steepness will make the correspondence between the upper and lower limits become "fat" and have a certain "thick tail", so the steepness can be appropriately fine-tuned manually to obtain the correspondence between the upper and lower limits that matches their massage intensity preference and willingness.
[0111] When the user has a need for a higher massage force with less restriction during the kneading process based on the real-time massage force graph, he or she can get corresponding visual guidance - increasing the translation amount will raise the upper limit value correspondence as a whole, and the user can manually fine-tune the translation amount appropriately to obtain the upper and lower limit value correspondence that matches his or her massage force preference and intention.
[0112] Example
[0113] Before leaving the factory, 120 volunteers of various ages and body shapes were randomly recruited, and they were given kneading massage techniques that just matched their waist massage intensity preferences, and data such as the rotation position of the massage head and massage intensity were collected.
[0114] The rotation position of the two massage heads when they are about to move from the farthest relative distance to the closest relative distance is -30, and the rotation position when they are about to reach the farthest relative distance is 30. In this way, the rotation position of the massage heads will go back and forth between -30 and 30 for each rotation. This rotation trajectory of the massage heads is to imitate the "kneading" technique. Each rotation of the massage heads is equivalent to completing a "kneading" technique cycle.
[0115] like Figure 3 The figure shows the data scatter density diagram of massage intensity with respect to the rotation position of the massage head, where the darker the area, the greater the density of the data scatter points, and the corresponding relationship between the massage intensity and the rotation position of the massage head (solid line in the figure). On this basis, the upper limit value corresponding relationship (dotted horizontal line in the figure) and lower limit value corresponding relationship (dashed line in the figure) that match the waist massage intensity preference of the volunteer group are obtained by adjusting the parameters, that is, the initial upper limit value curve and the initial lower limit value curve.
[0116] by Figure 3 Taking the initial upper limit curve and the initial lower limit curve as examples, the adjustment effects of the control parameters such as height, steepness, offset, and translation are as follows: Figure 4-Figure 7 As shown. Among them, Figure 4 In order to reduce the height control parameter, the value of a1 is reduced from "equal to 1" to "greater than 0 and less than 1". It can be seen that the shape of the corresponding relationship after adjustment is "shorter" than before adjustment, making the change in the massage force limit before and after the massage head rotation position is adjusted between -10 and 10 more significant.
[0117] Figure 5 In order to increase the steepness control parameter, the value of b1 is increased from "equal to 1" to "greater than 1". It can be seen that the shape of the corresponding relationship after adjustment is steeper than before adjustment, making the change in the massage intensity limit before and after adjustment when the massage head rotation position is between -20 to -10 (i.e. the middle section of squeezing) and 10 to 20 (i.e. the middle section of stretching) more significant.
[0118] Figure 6 To increase the offset control parameter, the value of c1 increases from "equal to 0" to "greater than 0". It can be seen that the corresponding relationship after adjustment is overall offset to the right (along the positive direction of the massage head rotation position axis) compared to before adjustment.
[0119] Figure 7 In order to reduce the translation control parameter (the value of d1 is reduced from "equal to 0" to "less than 0"), it can be seen that the corresponding relationship after adjustment is overall translated downward (along the negative direction of the massage force axis) compared to before adjustment.
[0120] Furthermore, the control parameters can be adjusted accordingly through several combinations, such as Figure 8 The figure shows the combined adjustment effect of height and translation. Figure 8 In order to increase the height and reduce the translation control parameters, the value of a1 is increased from "equal to 1" to "greater than 1" and the value of d1 is reduced from "equal to 0" to "less than 0", and the sum of a1 and d1 remains unchanged before and after the adjustment. It can be seen that the shape of the corresponding relationship after the adjustment is more "fat-tailed" than before the adjustment, making the change in the massage intensity limit before and after the adjustment when the massage head rotation position is between -30 and -20 (i.e., the end of the extrusion stage) and 20 to 30 (i.e., the end of the stretching stage) is more significant.
[0121] In the early stage of using the massage movement of the present invention, based on Figure 3 The initial upper and lower limit values correspond to each other, and a single experiential massage is performed on the waist of a specific user of the massage chair, and data such as the rotation position of the massage head and massage intensity are collected.
[0122] like Fig. 9 Shown are the data scatter points of massage intensity with respect to the rotation position of the massage head and the upper and lower limit curves (dark dotted horizontal lines in the figure) obtained through autonomous learning to match the massaged area with the massage intensity preference of a specific user.
[0123] During the whole process of using the massage movement, the graph drawn by the corresponding relationship between the actual value of the massage intensity of the massaged part and its upper and lower limits and the rotation position of the massage head within a certain massage time window can be directly displayed on the display screen, such as Fig.10 shown.
[0124] During the whole process of use, users can Fig.10 Select and modify the "correspondence" and "control parameters" on the upper left; for example, if the "upper limit correspondence" and "height" are selected respectively, the user can fine-tune the height, and the correspondence of the graph on the right will also change accordingly; the user then considers whether to accept the fine-tuning results of the selected "correspondence" and "control parameters" based on the changes in the correspondence in the graph on the right. Since the detected massage intensity can be displayed in real time on the display screen, the user can accurately judge how to adjust the massage force according to the massage force felt. The direction and amplitude of the adjustment can be easily selected on the display screen. After the adjustment is completed on the display screen, the user can immediately feel the change in massage intensity, thus forming a closed loop of massage intensity adjustment, so that the user can accurately and quickly adjust the massage intensity.
[0125] The above is only a specific implementation of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantial changes to the present invention using this concept shall be deemed as an infringement of the protection scope of the present invention.
Claims
1. A 5D movement self-learning massage method, characterized in that: include: The massage movement drives the massage head on the massage swing arm to apply a roughly circular rotation massage to a part of the user's body, with each circular rotation massage as a cycle, and the force sensor detects the pressure of the user's body on the massage head as the massage force; The massage head is moved to other parts of the user's body, and the massage is continued to be rotated in a roughly circular trajectory, with each circular rotation massage as a cycle, and the force sensor is used to detect the pressure exerted by the user's body on the massage head; The position sensor is used to detect the spatial position, i.e. the rotational position, of the massage head during the massage action in real time; The massage force data and the rotation position data detected by the force sensor and the position sensor are transmitted to the central processing unit, and the central processing unit is used to process and generate massage force curves of different parts of the user's body in each rotation massage cycle; The central processing unit controls the massage movement to push out or withdraw the massage head in the direction toward the user's body, so as to adjust the massage intensity applied by the massage head to the body parts in different rotation massage cycles; When the massage movement is initially used, the massage technique is performed, and the upper and lower limit curves reflecting the user's preferences for different body parts are obtained through autonomous learning based on the massage force data and rotation position data collected in real time; When the massage movement is subsequently used, the central processing unit controls the massage head to perform massage techniques within a range enclosed by the upper and lower limit curves corresponding to the preference for different parts of the human body.
2. The 5D movement self-learning massage method as claimed in claim 1, characterized in that: Also includes: Identify a specific user; When the massage movement is initially used, massage techniques are performed on a specific user whose identity has been identified based on the massage force curve within the rotational massage cycle corresponding to different body parts. Autonomous learning is performed based on the massage force data and rotation position data collected in real time to obtain upper and lower limit curves that reflect the preferences of different body parts of the specific user.
3. A 5D movement self-learning massage method as claimed in claim 1, characterized in that: The central processing unit uses nonlinear regression to establish a corresponding relationship curve between the massage force and rotation position of the massage head based on the data detected by the force sensor and the position sensor, and obtains the massage force curves of different parts of the human body in each rotation massage cycle after adjusting the parameters.
4. A 5D movement self-learning massage method as claimed in claim 3, characterized in that: The massage force curve includes an initial upper limit value curve and an initial lower limit value curve: f1=a1*h1(b1*(s-c1))+d1; f2=a2*h2(b2*(s-c2))+d2; Among them, f1 and f2 are the upper limit and lower limit of the massage intensity respectively, h1(.) and h2(.) represent the kernel functions of the initial upper limit curve and the initial lower limit curve respectively, a1 is the height control of the initial upper limit curve, a2 is the height control of the initial lower limit curve, b1 is the steepness control of the initial upper limit curve, b2 is the steepness control of the initial lower limit curve, c1 is the offset control of the initial upper limit curve, c2 is the offset control of the initial lower limit curve, d1 is the translation control of the initial upper limit curve, and d2 is the translation control of the initial lower limit curve.
5. A 5D movement self-learning massage method as claimed in claim 2, characterized in that: The different parts of the human body include shoulders, waist and buttocks, which correspondingly include the massage intensity curve and the preferred upper and lower limit curves of the shoulders in each rotational massage cycle, the massage intensity curve and the preferred upper and lower limit curves of the waist in each rotational massage cycle, and the massage intensity curve and the preferred upper and lower limit curves of the buttocks in each rotational massage cycle.
6. A 5D movement self-learning massage method as claimed in claim 2, characterized in that: The specific calculation method of the upper and lower limit curves reflecting the preferences of different body parts of the specific user is as follows: Divide the collected rotation position data into several intervals, and obtain the maximum and minimum values of the massage intensity data in each interval; Calculating the sum of distances between the maximum values in a plurality of the intervals and the corresponding upper limit values of the massage intensity curve, and the sum of distances between the minimum values in a plurality of the intervals and the corresponding lower limit values of the massage intensity curve; A genetic algorithm is used for feedback learning and autonomous adjustment to obtain a set of control parameters of the massage intensity curve so as to minimize the sum of the distances, and the massage intensity curve with the control parameters is used as the upper and lower limit curves of the preference.
7. A 5D movement self-learning massage method as claimed in claim 2, characterized in that: When the central processing unit determines that the real-time massage intensity exceeds the upper and lower limit curve range of the preference, the central processing unit controls the massage head on the massage movement to push out or withdraw so that the real-time massage intensity is within the upper and lower limit curve range of the preference; or, when a specific user actively increases pressure on the massage head, and the central processing unit determines that the real-time massage intensity exceeds the upper and lower limit curve range of the preference, the central processing unit controls the massage head on the massage movement to push out or withdraw so that the real-time massage intensity is within the upper and lower limit curve range of the preference.
8. A 5D movement self-learning massage method as claimed in claim 2, characterized in that: It also includes adjusting the strength of the massage head according to the expectations of a specific user, specifically including: The massage movement analyzes the adjustment direction, frequency and rate of change of the massage intensity before adjustment of the massage head's ejection stroke to continuously detect whether the user has an avoiding action or a pushing action. If there is an avoiding action or a pushing action, it means that the user expects to adjust the massage intensity of the massage head.
9. A 5D movement self-learning massage method as claimed in claim 8, characterized in that: The avoiding action refers to the behavior of the user lifting the massaged body part relative to the massage movement to reduce or even avoid contact with the massage head, and the avoiding action is regarded as the user's expectation to obtain a relatively smaller massage intensity; the pushing action refers to the behavior of the user pressing back on the massaged body part to push back the massage head in an antagonistic manner to increase contact with the massage head, and the pushing action is regarded as the user's expectation to obtain a relatively greater massage intensity.
10. A 5D movement self-learning massage method as claimed in claim 8, characterized in that: If the user's desire to adjust the massage intensity of the massage head is still detected after the intensity of the massage head is adjusted according to the user's desire, the control parameters of the preferred upper and lower limit curves are controlled to match the massage intensity preferred by the user.
11. A massage chair, comprising a massage movement and a central processing unit, characterized in that: A 5D movement self-learning massage method according to any one of claims 2 to 10 is adopted.
12. A massage chair as claimed in claim 11, characterized in that: The device also includes a display module, which displays in real time the massage intensity applied by the massage head to the body parts during different rotation massage cycles and displays the upper and lower limit curves of the preference.
13. The massage chair according to claim 11, characterized in that: It also includes an adjustment module, which is used to adjust the real-time massage force and the speed of the massage head movement, thereby changing the shape of the massage force curve; the adjustment module also allows the user to manually adjust the control parameters of the preferred upper and lower limit curves.
