Massage robot and its control method, control device and storage medium

Abstract

This disclosure provides a massage robot and its control method, control device, and storage medium. The massage robot control method of this disclosure includes: determining a massage path based on a user's human body point cloud data to obtain the user's massage path data, the massage path data including a massage movement trajectory and a first position, a first tolerance range, and a massage intensity range for each acupoint on the massage movement trajectory; controlling a robotic arm to perform multi-point light touches in a local area of ​​the first position of the acupoint and obtaining the pressure value of each light touch point; determining a second position of each acupoint on the massage movement trajectory based on the pressure value of each light touch point; and controlling the robotic arm to perform a massage operation according to the massage movement trajectory, the second position of each acupoint on the massage movement trajectory, the pre-set second tolerance range, and the massage intensity range. This disclosure can effectively improve the accuracy of acupoint positioning and the stability and safety of the massage effect.

Description

Technical Field

[0001] This disclosure relates to the field of robotics, and more particularly to a massage robot and its control method, control device and storage medium. Background Technology

[0002] Although home massage devices have become widely available, existing home massage devices still have many limitations. Currently, most roller or airbag massage chairs use preset fixed tracks or large-area coverage operations, lacking the ability to adapt to individual differences in user body shape. The accuracy of acupoint positioning is insufficient, making it difficult to apply precise pressure to key acupoints. The massage effect exhibits significant individual differences, and may even deviate from the target area, making it difficult to guarantee the stability and consistency of the massage effect, resulting in a poor user experience. Summary of the Invention

[0003] In view of this, the present disclosure provides a massage robot and its control method, control device and storage medium.

[0004] According to a first aspect of this disclosure, a method for controlling a massage robot is provided, the method comprising:

[0005] Determining a massage path includes: acquiring the user's human body point cloud data, determining a massage path based on the user's human body point cloud data to obtain the user's massage path data, the massage path data including a massage movement trajectory and the first position, first tolerance range, and massage intensity range of each acupoint on the massage movement trajectory;

[0006] Determining the second position of each acupoint on the massage movement trajectory includes: controlling the robotic arm to perform multi-point light touches in a local area of ​​the first position of the acupoint and obtaining the pressure value of each light touch point; determining the second position based on the pressure value of each light touch point; the local area of ​​the first position is determined based on the first position and the first tolerance range; and the second position is the position of the light touch point with the smallest change in pressure value in its neighborhood among all the light touch points.

[0007] Performing a massage operation includes controlling a robotic arm to perform a massage operation according to the massage movement trajectory and the second position of each acupoint on the massage movement trajectory, a pre-set second tolerance range and the massage intensity range, wherein the second tolerance range is smaller than the first tolerance range.

[0008] In some embodiments of the first aspect of this disclosure, before determining the massage path based on the user's human body point cloud data to obtain the user's massage path data, the method further includes: detecting the pressure difference between the left and right sides of the backrest when the user's back is in contact with the backrest of the massage robot in a sitting posture, and adjusting the tilt of the backrest of the massage robot according to the pressure difference between the left and right sides of the backrest so that the pressure difference between the left and right sides of the backrest is less than or equal to a preset first pressure difference threshold.

[0009] In some embodiments of the first aspect of this disclosure, determining the massage path based on the user's human body point cloud data to obtain the user's massage path data includes: constructing a virtual skeletal model of the user using the user's human body point cloud data; performing acupoint mapping based on the user's virtual skeletal model and a pre-configured acupoint feature database to determine the user's massage path and generating the massage path data.

[0010] In some embodiments of the first aspect of this disclosure, the step of determining the massage path based on the user's human body point cloud data to obtain the user's massage path data further includes: detecting the deviation between the user's body center of gravity and the pressure center of gravity of the massage robot in a sitting posture and adjusting the virtual skeleton model according to the deviation so that the virtual skeleton is consistent with the user's body posture.

[0011] In some embodiments of the first aspect of this disclosure, performing the massage operation includes: when the deviation between the massage force of the massage operation and the massage force range of the corresponding acupoint exceeds a preset massage force deviation range, adjusting the position of the robotic arm until the deviation between the massage force of the massage operation and the massage force range of the corresponding acupoint is less than a preset deviation threshold.

[0012] In some embodiments of the first aspect of this disclosure, the massage operation further includes: acquiring the user's electroencephalogram (EEG) signal, electrodermal signal, heart rate signal, facial expression image, skin temperature and / or user instructions during the massage operation, and adjusting the massage parameters of the massage operation according to the user's EEG signal, electrodermal signal, facial expression image, skin temperature and / or operation instructions, wherein the massage parameters include the second location of the acupoint, the duration of massage stay and / or the massage intensity.

[0013] In some embodiments of the first aspect of this disclosure, the massage operation further includes: during the massage operation, acquiring the user's electroencephalogram (EEG), electrodermal (EDS), heart rate, skin temperature, and / or facial expression images, and periodically assessing the user's massage comfort based on the EEG, EDS, heart rate, skin temperature, and / or facial expression images during the massage; if the user's massage comfort is lower than a preset comfort threshold after N consecutive assessments, the step of determining the massage path is re-executed, where N is an integer greater than 1 and a preset value is taken.

[0014] According to a second aspect of this disclosure, a control device for a massage robot is provided, the control device for the massage robot comprising:

[0015] The coarse adjustment module is used to determine the massage path. The determination of the massage path includes: acquiring the user's human body point cloud data, and determining the massage path based on the user's human body point cloud data to obtain the user's massage path data. The massage path data includes the massage movement trajectory and the first position, first tolerance range, and massage intensity range of each acupoint on the massage movement trajectory.

[0016] The fine-tuning module is used to determine the second position of each acupoint on the massage movement trajectory. Determining the second position of each acupoint on the massage movement trajectory includes: controlling the robotic arm to perform multi-point light touches in a local area of ​​the first position of the acupoint and obtaining the pressure value of each light touch point; determining the second position based on the pressure value of each light touch point; the local area of ​​the first position is determined based on the first position and the first tolerance range; and the second position is the position of the light touch point with the smallest change in pressure value in its neighborhood among all the light touch points.

[0017] The massage operation module is used to perform massage operations, which include controlling the robotic arm to perform massage operations according to the massage movement trajectory and the second position of each acupoint on the massage movement trajectory, the preset second tolerance range and the massage intensity range, wherein the second tolerance range is smaller than the first tolerance range.

[0018] According to a third aspect of this disclosure, a massage robot is provided, the massage robot comprising: a control unit, a 3D vision sensor, a robotic arm, and a robotic hand, wherein a second pressure sensor is disposed on the robotic hand, the 3D vision sensor, the robotic arm, and the second pressure sensor are respectively connected to the control unit, and the robotic hand is connected to the robotic arm; wherein the control unit comprises: one or more processors and a memory storing programs, the programs comprising instructions, which, when executed by the processor, cause the processor to perform the above-described method.

[0019] According to a fourth aspect of this disclosure, a computer-readable storage medium storing a program is provided, the program including instructions that, when executed by one or more processors of a computing device, cause the computing device to perform the method described above.

[0020] As can be seen from the above technical solutions, the embodiments disclosed herein can effectively improve the accuracy of acupoint positioning, realize automatic massage based on dynamic adaptation and personalized precise adjustment of the massage robot, effectively improve the stability and safety of the massage effect, and thus effectively improve the user experience of the massage robot. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart of a control method for a massage robot provided in an embodiment of this disclosure;

[0023] Figure 2 This is a schematic diagram of the massage process of the massage robot provided in an embodiment of the present disclosure;

[0024] Figure 3 This is a schematic diagram of the structure of the massage robot control device provided in the embodiments of this disclosure;

[0025] Figure 4 A schematic structural block diagram of a massage robot provided in an embodiment of this disclosure. Detailed Implementation

[0026] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0027] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of this disclosure. The singular forms “a,” “the,” and “the” as used in the embodiments of this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0028] Depending on the context, words such as "if," "when," etc., used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrases "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0029] As mentioned in the background section, existing massage devices suffer from insufficient accuracy in acupoint positioning, resulting in a lack of personalized massage effects and a poor user experience.

[0030] In addition, existing massage devices have at least the following drawbacks:

[0031] First, it lacks the ability to adapt to changes in user posture, making it difficult for acupoints to be located in response to these changes.

[0032] Secondly, existing massage devices rely heavily on manual selection by the user in terms of control modes, lacking a real-time perception and feedback mechanism for the user's physiological state. This can easily lead to overload or under-stimulation, affecting the massage effect and posing a risk to comfort.

[0033] Third, existing massage devices have relatively simple interaction methods, limited to switching preset programs and simple intensity adjustments, making it difficult to support users in making delicate and continuous personalized parameter adjustments during the massage process.

[0034] Fourth, existing massage devices lack a closed-loop control strategy based on real-time physiological signals. When encountering sudden discomfort or abnormal body position of the user, the response is delayed, which poses certain risks and poor safety.

[0035] In view of this, the present disclosure provides the following massage robot and its control method, control device and storage medium, which adopts a hierarchical control framework. That is, the massage process of the massage robot is divided into two stages: coarse adjustment and fine adjustment. The initial position of the acupoint is determined by planning the massage path through coarse adjustment (i.e., the first position in this document). Fine adjustment further accurately locates the position of the acupoint by using the pressure signal of the robotic arm contacting the acupoint area within the first tolerance range provided by coarse adjustment (i.e., the second position in this document). Finally, the massage operation is performed on the acupoint area defined by the second position and the second tolerance range. Thus, the present disclosure realizes automatic massage based on dynamic adaptation and personalized precise adjustment of the massage robot through such a hierarchical control framework of coarse adjustment and fine adjustment. The acupoint positioning accuracy is high, which can effectively improve the massage effect and user experience of various users.

[0036] Furthermore, the embodiments disclosed herein can also optimize massage parameters in real time by combining the fusion feedback of multi-source signals such as EEG, EEG, and heart rate during the fine-tuning stage, and return to the coarse-tuning stage to correct the massage path when necessary, thereby realizing a real-time perception and feedback mechanism of the user's physiological state and multi-level closed-loop control, effectively improving comfort and safety.

[0037] It should be noted that the location of the acupoints involved in the embodiments of this disclosure (e.g., the first location, the second location) can be represented by coordinates in the machine coordinate system of the massage robot. The machine coordinate system is a three-dimensional Cartesian coordinate system with three mutually perpendicular axes (X, Y, Z). The location of any point of the massage robot relative to the user's body can be represented as three-dimensional coordinates (x, y, z), where x represents the X-axis coordinate of the location, y represents the Y-axis coordinate of the location, and z represents the Z-axis coordinate of the location.

[0038] The specific implementation methods of the massage robot control method according to the present disclosure will be described in detail below.

[0039] Figure 1 A flowchart illustrating a massage robot control method provided in an embodiment of this disclosure is shown. This control method can be executed by a control unit in the massage robot. See also... Figure 1 The control methods for massage robots include:

[0040] Step 101, determine the massage path, which includes: acquiring the user's human body point cloud data, determining the massage path based on the user's human body point cloud data to obtain the user's massage path data, the massage path data including the massage movement trajectory and the first position, first tolerance interval and massage intensity interval of each acupoint on the massage movement trajectory;

[0041] Step 102, determine the second position of each acupoint on the massage movement trajectory. Determining the second position of each acupoint on the massage movement trajectory includes: controlling the robotic arm to perform multi-point light touches in a local area of ​​the first position of the acupoint and obtaining the pressure value of each light touch point. The second position is determined based on the pressure value of each light touch point. The local area of ​​the first position is determined based on the first position and the first tolerance range. The second position is the position of the light touch point with the smallest change in pressure value in the neighborhood among all the light touch points.

[0042] Step 103, perform massage operation. Performing massage operation includes controlling the robotic arm to perform massage operation according to the massage movement trajectory and the second position of each acupoint on the massage movement trajectory, the pre-set second tolerance range and the massage intensity range. The second tolerance range is smaller than the first tolerance range.

[0043] The method of this embodiment first determines the massage movement trajectory and the initial position of each acupoint (i.e., the first position) based on the user's human body point cloud data to achieve coarse adjustment. Then, based on the initial position and initial tolerance range of each acupoint (i.e., the first tolerance interval), it determines the fine position of the acupoint (i.e., the second position) to achieve fine adjustment of the acupoint positioning position. Subsequently, it performs massage operation according to the massage movement trajectory and the fine position of each acupoint on it and the tolerance interval of fine massage (i.e., the second tolerance interval). The massage intensity of the massage operation is controlled within the predetermined massage intensity range of the acupoint. Thus, adaptive and precise massage is achieved, the acupoint positioning accuracy is high and closer to the real human acupoint sensitive points, and the massage intensity can be controlled within a reasonable range, which can effectively improve the massage effect and user experience of different users.

[0044] Furthermore, considering that a user's posture on the massage chair may change slightly, such as shoulders slumping or body tilting, the method of this embodiment may further include, before step 101: detecting the pressure difference between the left and right sides of the massage robot backrest when the user's back is in contact with the backrest of the massage robot in a sitting posture, and adjusting the tilt of the massage robot backrest according to the pressure difference between the left and right sides of the backrest so that the pressure difference between the left and right sides of the backrest is less than or equal to a preset first pressure difference threshold. Thus, by comparing the pressure difference between the left and right sides of the backrest, the tilt of the massage robot backrest can be adjusted to ensure the user's sitting posture is balanced, avoiding a decrease in subsequent massage effect due to an unbalanced user sitting posture.

[0045] Specifically, the process of adjusting the backrest tilt of the massage robot by comparing the pressure difference between the left and right sides of the backrest can include the following steps a1 to a3:

[0046] Step a1: Obtain the first pressure array data generated by the interaction between the user's back and the backrest of the massage robot in a sitting posture;

[0047] Step a2: Determine the pressure difference between the left and right sides of the backrest based on the data from the first pressure array;

[0048] Specifically, the pressure sensor array deployed on the back of the massage robot is divided into left and right sections beforehand, forming a matrix corresponding to the left side of the backrest. Corresponding to the right side area of ​​the backrest The difference between the total pressure in the left side area and the total pressure in the right side area of ​​the backrest is calculated by the following formulas (1) to (3). The difference between the total pressure in the left side area and the total pressure in the right side area of ​​the backrest is the pressure difference between the left and right sides of the backrest.

[0049] (1)

[0050] (2)

[0051] (3)

[0052] in, This indicates the total pressure in the left side area of ​​the seatback. This represents the total pressure in the right-side area of ​​the backrest, and Σ represents the summation operation. This represents the pressure value at coordinates (x, y) in the matrix representing the left side of the backrest area. This represents the pressure value at coordinate (x, y) in the matrix representing the right side region of the backrest. This indicates the pressure difference between the left and right sides.

[0053] Step a3: Adjust the tilt of the massage robot backrest according to the pressure difference between the left and right sides of the backrest, so that the pressure difference between the left and right sides of the user's back is less than or equal to the preset first pressure difference threshold.

[0054] Specifically, if the absolute value of the pressure difference between the left and right sides of the backrest... If the pressure difference exceeds the preset first pressure difference threshold, the backrest will tilt towards the side with lower total pressure until the absolute value of the pressure difference between the left and right sides of the backrest is reached. The pressure difference converges to within the first pressure difference threshold range (i.e., the pressure difference between the left and right sides of the backrest is less than or equal to the first pressure difference threshold). If the absolute value of the pressure difference between the left and right sides of the backrest, |ΔF|, is less than or equal to the first pressure difference threshold, then the user's current sitting posture is determined to be a balanced sitting posture, and there is no need to adjust the backrest tilt.

[0055] The first pressure difference threshold can be pre-configured. In practical applications, different pressure difference thresholds can be tested for different users and saved in advance. In some examples, the specific value of the first pressure difference threshold can be determined through testing. For example, the first pressure difference threshold can be set to 5% to 10% of the total pressure on one side of the backrest, where the total pressure on one side of the backrest refers to the total pressure obtained from testing. In other examples, the first pressure difference threshold can also be an empirical value obtained in advance through testing, which can be a fixed value. Therefore, the massage robot can automatically adjust the tilt angle of the left and right sides of the backrest by comparing the pressure difference between the left and right sides of the user's back to ensure the user is in a balanced sitting posture.

[0056] Before adjusting the backrest of the massage robot, the headrest can also be automatically adjusted to keep the user's head and neck stable and balanced, ensuring more accurate adjustment of the massage robot's backrest tilt. Specifically, the massage robot's headrest can be driven up and down by a stepper motor, pausing for 1 second at each position. Pressure sensors on the headrest monitor the pressure in real time. When the pressure on the headrest is concentrated in the middle and the total pressure does not exceed the standard, that position can be locked as the initial neck support point.

[0057] In step 101, there are various ways to determine the massage path. Any method applicable to the embodiments of this disclosure can be used to implement the process of determining the massage path in step 101. The massage path is determined using human point cloud data. The goal is to quickly and macroscopically locate the user's skeletal position and major acupoints, generating massage path data. This massage path data includes the preliminary location of the acupoints, i.e., the first position.

[0058] In some implementations, step 101, determining the massage path, may include: constructing a virtual skeletal model of the user using the user's human body point cloud data; and performing acupoint mapping based on the user's virtual skeletal model and a pre-configured acupoint feature database to determine the user's massage path and generate massage path data. Determining the massage path through the virtual skeletal model and acupoint mapping can dynamically adapt to the user's individual body shape to initially locate acupoints, effectively improving the accuracy of initial acupoint location.

[0059] In practical applications, when a user sits in the massage chair of the massage robot, multiple sensors distributed on the chair body start working. The 3D vision sensors mounted on the massage robot collect the user's human body point cloud data (that is, the user's whole body point cloud data or point cloud data of a specified part) and provide it to the control unit. Based on the user's human body point cloud data, the control unit uses human posture estimation algorithms to identify key points such as acromion, sternum, navel, hip bone, etc., and constructs a virtual skeletal model of the user through these key points.

[0060] A pre-configured acupoint feature database can be used, which can contain acupoint feature data corresponding to different heights. By mapping the acupoint feature data corresponding to the current user's height in the acupoint feature database with the user's virtual skeletal model, the first position of each acupoint can be determined.

[0061] The first tolerance range can be preset. Specifically, a tolerance range can be pre-marked for each acupoint. When the primary location of an acupoint is determined, its first tolerance range can be obtained by querying its marking information. In some examples, the first tolerance range is related to the corresponding body part of the acupoint. If the acupoint is in the neck region, its first tolerance range is the preset neck region tolerance range; if the acupoint is in the lower back region, its first tolerance range is the preset lower back region tolerance range. For example, the neck region tolerance range can be set to ±8 mm, and the lower back region tolerance range can be set to ±12 mm. In practical applications, the first tolerance range for each acupoint and the tolerance ranges for different body parts can be preset or dynamically adjusted as needed.

[0062] The massage movement trajectory is related to the pre-selected massage mode. For example, if the currently selected massage mode is the shoulder and neck relaxation mode, then after completing acupoint mapping, the massage movement trajectory is generated according to the pre-configured top-down massage path. If the currently selected massage mode is the lower back soothing mode, then after completing acupoint mapping, the massage movement trajectory is generated according to the strategy of "massage path bidirectionally symmetrically distributed along both sides of the spine".

[0063] The massage intensity range for acupoints is related to the corresponding body part. If an acupoint is located in the neck area, its massage intensity range is defined as the neck massage intensity range; if the acupoint is located in the shoulder and back area, its massage intensity range is defined as the shoulder and back massage intensity range; and if the acupoint is located in the lower back and hip area, its massage intensity range is defined as the lower back and hip massage intensity range. In practical applications, the massage intensity ranges for the neck, shoulder and back, and lower back and hip areas can be preset or dynamically adjusted as needed. For example, the neck massage intensity range can be set to 12-25 Newtons, the shoulder and back massage intensity range to 20-35 Newtons, and the lower back and hip massage intensity range to 25-40 Newtons.

[0064] The local area of ​​the first position can be an area centered on the first position of the acupoint and with a diameter within the first tolerance range of the acupoint. For example, if the acupoint is in the neck region, its first tolerance range is a preset neck region tolerance range, which can be set to ±8 mm. Therefore, the local area of ​​the first position can be an area centered on the first position of the acupoint and with a diameter of 8 mm.

[0065] As described above, step 101 can use a 3D vision sensor to collect the user's human body point cloud data, virtual skeleton and acupoint database to determine the massage movement trajectory, and lock the positioning range of each acupoint on the massage movement trajectory in the area defined by the first position and the first tolerance interval. This area is an area with a diameter of about 5~10mm, realizing the coarse adjustment of the massage path and acupoint position that dynamically adapts to the user's individual body shape.

[0066] Considering that a user's posture on a massage chair may change slightly, such as shoulders slumping or body tilting, this can cause inconsistencies between the virtual skeletal model and the user's actual body posture, resulting in deviations in acupoint mapping and affecting the accuracy of massage position positioning. Step 101 may further include: detecting the deviation between the user's center of gravity and the massage chair's pressure center of mass in the sitting posture and automatically adjusting the virtual skeletal model based on this deviation, so that the virtual skeleton is consistent with the user's body posture, reducing acupoint mapping deviations and further improving the accuracy of massage position positioning.

[0067] An exemplary implementation process for detecting the deviation between the user's center of gravity and the pressure center of the massage chair in a sitting posture and automatically adjusting the virtual skeletal model based on this deviation may include the following steps b1 to b5:

[0068] Step b1: Determine the user's center of gravity position using the user's human body point cloud data;

[0069] Step b2: Obtain the second pressure array data generated by the user's lower back contacting the massage robot backrest and the buttocks and legs contacting the massage robot seat when in a sitting posture, and determine the position of the pressure centroid based on the second pressure array data;

[0070] Specifically, the second pressure sensor array on the backrest and seat of the massage robot records the second pressure array data in real time and provides it to the control unit. This second pressure array data can indicate the distribution of pressure generated by the user's body contacting and interacting with the corresponding parts of the massage robot in a sitting posture. The control unit uses the second pressure array data to determine the position of the pressure center of mass.

[0071] The second pressure array data is a two-dimensional matrix. Each element in this matrix represents the pressure value (in N) collected by a pressure sensor in the second pressure sensor array at its location. The value of each element in the second pressure array data can be obtained by converting the raw pressure signal collected by the corresponding pressure sensor in the second pressure sensor array through methods such as calibration curves. In specific applications, each pressure sensor is configured with its own unique "pressure-voltage" correspondence at the factory; this "pressure-voltage" correspondence is the calibration curve.

[0072] Specifically, the position of the pressure centroid can be calculated using the following formula (4):

[0073] (4)

[0074] in, and This represents the coordinates of the center of mass in the pressure sensor array coordinate system. and Indicates pressure point The element position in the second pressure array data. Indicates pressure point Pressure value, =1,2,3,……,n, where n represents the total number of elements in the second pressure array data, and ∑ represents summation.

[0075] Step b3: Calculate the distance between the user's center of gravity and the pressure center of gravity on the same reference plane. This distance is the deviation between the user's center of gravity and the pressure center of gravity of the massage robot.

[0076] Step b4: If the distance between the user's center of gravity and the pressure center of gravity on the same reference plane is greater than the first distance threshold, the virtual skeleton model is corrected for posture.

[0077] Step b5: If the distance between the user's center of gravity and the pressure center of gravity on the same reference plane is less than or equal to the first distance threshold, then stop the posture correction of the virtual skeleton model, or do not perform any action, that is, there is no need to perform posture correction of the virtual skeleton model.

[0078] The first distance threshold can be pre-configured. For example, the first distance threshold for people of different heights can be pre-tested and saved in the massage robot. Before use, users can test the first distance threshold to find one that suits them or query the massage robot for the first distance threshold corresponding to their height. Alternatively, the first distance threshold can be set to a fixed value, which is determined through testing and pre-saved in the massage robot's control unit. For example, the first distance threshold can be set to 8 millimeters or other values.

[0079] Assuming the first distance threshold can be set to 8 mm, if the distance between the user's center of gravity and the pressure center of gravity on the same reference plane exceeds 8 mm, the virtual skeleton model will automatically be triggered to correct its posture until the distance between the user's center of gravity and the pressure center of gravity on the same reference plane is within 8 mm. This ensures that the virtual skeleton model is consistent with the user's actual sitting posture and effectively improves the accuracy of acupoint positioning.

[0080] In step 102, the control unit of the massage robot sends a command to the robotic arm to control the robotic arm to send the robotic hand to the local area of ​​the first position of the acupoint determined in step 101. The diameter of the local area is the first tolerance range. In the local area of ​​the first position of the acupoint, the pressure generated by the contact is sampled by the second pressure sensor deployed at the end of the robotic hand through fine contact and pressure distribution analysis is performed. In the neighborhood range (e.g., within ±5 mm), the second position of the acupoint is determined by scanning multiple points and finding the point with the most uniform pressure distribution. Thus, the fine positioning of each acupoint on the massage movement trajectory is realized.

[0081] Through the processing in step 102, the massage operation area of ​​a certain acupoint can be locked in a local area of ​​the second position. The local area of ​​the second position can be an area with a diameter of the second position as its center and a second tolerance interval. The second tolerance interval is smaller than the first tolerance interval, and the size of the second tolerance interval is determined by the range of the sensitive point of the acupoint. For example, it is known from experience that the sensitive point of the human acupoint is often within a range of 2 to 3 millimeters in diameter. Therefore, the second tolerance interval can be set to 2 to 3 millimeters. Correspondingly, the massage operation area for a certain acupoint, that is, the local area of ​​the second position, can be an area with a diameter of about 2 to 3 millimeters centered on the second position. It can be seen that step 102 not only further refines the positioning position of the acupoint through the pressure signal generated by multi-point light touch, realizing fine adjustment of the positioning position of each acupoint on the massage movement path, but also locks the positioning range of the acupoint to conform to the size of the sensitive point of the human acupoint. The size of the massage operation area for a certain acupoint can also better conform to the actual sensitive point range of the human acupoint, solving the problem that a slight deviation when the acupoint massage operation area is locked within the range of the human sensitive point may lead to a weakening of the massage effect or even discomfort.

[0082] Furthermore, in step 102, the specific implementation process of determining the second position may include the following steps c1 to c2:

[0083] Step c1: Select candidate points from the various touch points;

[0084] In some examples, candidate points can be selected using the average and standard deviation of the pressure values ​​of all touch points. Specifically, the average μ and standard deviation σ of the pressure values ​​of all touch points can be calculated, and points whose pressure values ​​fall within the interval [μ - kσ, μ + kσ] determined by the average and standard deviation can be selected as candidate points, where k is a constant, k = 0.5 or 1. In other examples, clustering algorithms such as K-means or other similar algorithms can be used to select candidate points from the touch points. This disclosure does not limit the specific method used to select candidate points.

[0085] In some examples, a candidate point can be any point among all touch points that meets preset conditions. These preset conditions may include: the pressure value is within a predetermined contact force range and the pressure fluctuation around the touch point is less than a preset fluctuation threshold. For example, if the contact force range is preset to 2~3N and the fluctuation threshold is preset to 0.3N, if the pressure value of a certain touch point is 2.5N, and the difference between the pressure values ​​of the touch point and those of the other touch points around it (i.e., the pressure fluctuation around the touch point) does not exceed 0.3N, then the touch point can be considered a candidate point.

[0086] Step c2: Calculate the neighborhood pressure standard deviation of each candidate point, and select the candidate point with the smallest pressure standard deviation as the second position. The coordinates of this candidate point are the second position.

[0087] Specifically, the neighborhood can be defined by a first tolerance interval and a second tolerance interval. To improve accuracy while reducing computational complexity, the size of the neighborhood can be set to be larger than the second position tolerance interval and smaller than the first position tolerance interval. For example, assuming the first position tolerance interval is ±8 mm and the second position tolerance interval is ±2~±3 mm, the neighborhood diameter can be set to ±5 mm, and the neighborhood range is ±5 mm.

[0088] In some examples, step c2 may also include: if multiple candidate points have the same minimum pressure standard deviation, one of those candidate points may be selected as the second position.

[0089] In some examples, step c2 may further include: if multiple candidate points have the same minimum pressure standard deviation, a candidate point that meets one of the following conditions may be selected as the second location: 1) the absolute value of the pressure deviation between all touch points in the neighborhood of the candidate point and the candidate point is less than a preset first pressure threshold; 2) more than M% of the touch points in the neighborhood of the candidate point have an absolute value of pressure deviation between the candidate point and the candidate point that is less than the first pressure threshold. The first pressure threshold can be a fixed value, determined empirically; in other words, the second pressure threshold can be an empirical value.

[0090] In some examples, step c2 may also include: eliminating candidate points that do not meet the aforementioned conditions 1) and 2) before calculating the neighborhood pressure standard deviation of each candidate point; calculating the neighborhood pressure standard deviation for candidate points that meet one or more of the aforementioned conditions 1) and 2) and selecting the candidate point with the smallest pressure standard deviation as the second position.

[0091] Minimizing the standard deviation ensures the lowest overall fluctuation in pressure distribution (i.e., minimal change in overall pressure value). The absolute deviation threshold ensures there are no outliers within the local neighborhood of the candidate point, resulting in an absolutely uniform pressure distribution. Therefore, a second location with minimal pressure value change and an absolutely uniform pressure distribution within its neighborhood can be found. This second location is the center of a region that is "completely flat without any protrusions or depressions." In other words, the local area centered on this second location better represents the true sensitive points of acupoints in the human body.

[0092] Although the pressure standard deviation is used as the basis for selecting the second position in the foregoing embodiments, those skilled in the art should understand that, for example, variance, pressure range, etc., can also be used as the basis for selecting the second position. In this regard, the embodiments disclosed herein do not impose any limitations.

[0093] Furthermore, in step 102, during multi-point light touches, the pressure value of the robotic arm's touch is controlled within a preset contact force range to avoid discomfort that may be caused by excessive touch pressure and a decrease in fine-tuning accuracy that may be caused by insufficient touch pressure. In specific applications, the contact force range can be pre-configured based on various factors such as specific application scenarios and user needs. For example, the contact force range can be set to 2.0~3.0 Newtons.

[0094] In step 103, performing the massage operation may include: controlling the robotic arm to drive the robotic hand to perform massage actions such as continuous pressing, kneading, rubbing, or circular massage in a local area with the second acupoint as the core and the second tolerance range as the diameter.

[0095] In step 103, periodic repeated scanning can be performed when the user's sitting posture changes slightly, and the second position of the acupoint can be dynamically updated in the manner of step 102 to ensure that the massage effect is continuous, stable, comfortable and effective.

[0096] Furthermore, in step 103, the massage intensity can be flexibly adjusted as needed. Specifically, step 103 may also include: when the deviation between the massage intensity of the massage operation and the massage intensity range of the corresponding acupoint exceeds a preset massage force deviation range, adjusting the position of the robotic arm until the deviation between the massage intensity of the massage operation and the massage intensity range of the corresponding acupoint is less than a preset deviation threshold. Here, the deviation between the massage intensity of the massage operation and the massage intensity range of the corresponding acupoint can be the maximum value of the difference between the two endpoints of the massage intensity range. Thus, by comparing the massage intensity range and the real-time massage force, the massage position of the robotic arm and robotic hand can be corrected in real time, ensuring that the massage operation is continuous, stable, and safe, while further improving the massage effect.

[0097] The deviation threshold can be flexibly determined and pre-configured based on various factors such as the actual application scenario, massage mode, massage area, and user needs. For example, the deviation threshold can be ±2 Newtons.

[0098] A closed-loop force control algorithm can be used to adjust the position of the robotic arm based on the deviation between the massage intensity of the massage operation and the corresponding acupoint massage intensity range. Specifically, the massage intensity range is compared with the actual massage intensity in real time. When the deviation exceeds a deviation threshold, the control unit uses the closed-loop force control algorithm to correct the position of the robotic arm. For example, if the massage intensity range is less than 25 Newtons, while the massage intensity of the massage operation is 28 Newtons, the deviation is 3 Newtons, which exceeds the deviation threshold by about 1 Newton. At this time, the robotic arm can be controlled to retract outward by about 1 mm until the deviation is less than the deviation threshold. When controlling the position adjustment of the robotic arm, the entire force control loop can operate at a frequency of 500 Hz to ensure stable and safe operation.

[0099] Specifically, the closed-loop force control algorithm can be, but is not limited to, PID control, fuzzy control, reinforcement learning, adaptive control, etc., to improve the control accuracy and adaptability when adjusting the position of the robotic arm under complex conditions. This disclosure does not impose any limitations on this aspect.

[0100] Furthermore, step 103 may also include: during the massage operation, acquiring the user's EEG signals, EEG signals, heart rate signals, facial expressions, skin temperature, and / or user commands, and adjusting the massage parameters of the massage operation based on the user's EEG signals, EEG signals, facial expressions, skin temperature, and / or operation commands. Massage parameters may include, but are not limited to, the second location of acupoints, massage dwell time, and / or massage intensity. Thus, not only can the massage intensity be maintained within the target range through high-precision force control closed loops such as millisecond-level, but the massage parameters can also be adjusted in real time using second-level physiological signals such as EEG signals, EEG signals, heart rate signals, and skin temperature. Furthermore, the massage parameters can be adjusted in response to user commands, thereby further improving the massage effect and user experience.

[0101] Specifically, the EEG sensor records the user's scalp potential to collect the user's EEG signals and transmits them to the control unit. The control unit calculates the power ratio of alpha waves (8–13 Hz) to beta waves (13–30 Hz) in the EEG signals in real time (i.e., the α / β power ratio). If the α / β power ratio increases by a first predetermined proportion (e.g., 10%) of the α / β baseline value, it indicates that the user has entered a relaxed state, and the current intensity and massage path can be maintained. If the α / β power ratio decreases by a second predetermined proportion (e.g., 90%) of the α / β baseline value, the massage intensity can be automatically reduced proportionally (e.g., reduced by about 15%), and the dwell time can be shortened. At the same time, the user can be prompted so that the user can actively adjust their posture in a timely manner.

[0102] Alpha waves are brain waves with frequencies between 8 and 13 Hz. They typically appear when the eyes are closed, the brain is relaxed, and the patient is awake and at rest. Alpha waves are a sign that the brain is in a calm, introspective state, not focused on the external world. For example, alpha wave activity increases during meditation or quiet rest. Beta waves are brain waves with frequencies between 13 and 30 Hz. They typically dominate when the eyes are open, thinking, problem-solving, focusing, or experiencing anxiety or tension. Beta waves represent a highly active and excited state of the cerebral cortex. The alpha / beta power ratio can be used to measure the balance between the brain's "relaxed" and "excited" states.

[0103] The α / β baseline value can be monitored in advance. Specifically, the average α / β power ratio measured when the user is in a quiet, neutral state can be monitored in advance, and this average value can be used as the user's α / β baseline value. For each user, their personalized baseline value can be monitored and saved before the massage.

[0104] Specifically, during the user's massage, the skin conductance sensor monitors changes in the user's skin conductivity to collect skin electrical signals, which are the conductivity values. If the user's conductivity increases by more than a preset first conductivity threshold (e.g., 0.2 microseconds) within a first predetermined duration (e.g., within 10 seconds), the control unit controls the robotic arm to immediately reduce the massage intensity and switch to a gentle mode. If the user's conductivity changes by less than a preset second conductivity threshold (e.g., 0.05 microseconds) within the first predetermined duration (e.g., within 10 seconds), it indicates that the user is stably relaxed, and the duration of the massage can be extended proportionally (e.g., by approximately 20%).

[0105] Specifically, during the user's massage, the camera can capture images of the user's facial expressions. The facial expressions can be recognized. If the user's expression is characterized by furrowed brows or downturned corners of the mouth, it can be determined that the user is currently uncomfortable, and the robotic arm can be controlled to immediately reduce the massage intensity or massage trajectory. If the user's expression is characterized by a wide smile, a gentle smile, or upturned corners of the mouth, the robotic arm can be controlled to maintain the current massage intensity and massage duration.

[0106] The specific implementation process of adjusting massage parameters based on heart rate signals and skin temperature is similar to that of the aforementioned electroencephalogram (EEG) signals, electrodermal signals, and facial expression images, and will not be described in detail here.

[0107] In addition, users can actively adjust massage parameters such as massage position, massage intensity, and / or massage duration through various methods such as voice and touch. That is, user commands can be, but are not limited to, voice commands, touch commands, or any other type of user-issued commands.

[0108] In some examples, a user can issue a voice command like "lighter," and the massage robot will respond by decreasing the current massage intensity according to a pre-configured intensity adjustment range (e.g., 3 Newtons). Conversely, a user can issue a voice command like "upper," and the massage robot will respond by moving the coordinates of the second position upwards by 5 millimeters according to a pre-configured position adjustment range (e.g., 5 millimeters). In practical applications, voice recognition algorithms, such as keyword matching algorithms, can be used to achieve real-time adjustment of massage parameters based on user voice commands. This ensures that the massage robot can accurately execute user voice commands even in noisy environments.

[0109] In some examples, users can also make fine adjustments via the armrest panel. For instance, the human-computer interaction module of the massage robot can provide physical buttons such as intensity sliders, four-way position buttons, and / or mode switching buttons. Users can input touch commands through these physical buttons, and the control unit of the massage robot responds to the touch commands by adjusting the massage intensity, massage position, massage dwell time, and / or massage action execution mode (e.g., switching from kneading to rubbing) within preset safety boundaries (e.g., a maximum neck massage intensity of 30 Newtons). Thus, users can instantly correct massage parameters through voice, panel input, etc., achieving a closed-loop user interaction and further enhancing the user experience.

[0110] In addition to voice commands and touch commands, user commands can also be input through various other methods such as eye tracking, gesture recognition, and remote control via a mobile app. This disclosure does not limit the specific form of user commands or their input methods.

[0111] Furthermore, the method of this embodiment may further include: during the massage operation, acquiring the user's electroencephalogram (EEG) signal, electrodermal signal, heart rate signal, skin temperature and / or facial expression image, and periodically evaluating the user's massage comfort based on the user's EEG signal, electrodermal signal, heart rate signal, skin temperature and / or facial expression image during the massage; if the user's massage comfort is lower than a preset comfort threshold after N consecutive evaluations, then returning to re-execute step 101, where N is an integer greater than 1 and a preset value is taken.

[0112] The user's massage comfort level is periodically assessed based on EEG signals, dermal conductivity signals, heart rate signals, skin temperature, and / or facial expressions during the massage process (the massage comfort level is a value from 0 to 1). If the user's massage comfort level is lower than a preset comfort threshold (e.g., 0.5) after N consecutive assessments, the system automatically reverts to step 101 to redetermine the massage path. Here, N is a preset integer greater than 1. Therefore, a dual-layer closed-loop control system with coarse and fine adjustments ensures that the system can respond quickly and dynamically correct deviations.

[0113] Specifically, the α / β power ratio calculated based on EEG signals is less than a preset α / β power ratio threshold (e.g., 0.9), the increase in the user's conductivity within 10 seconds, based on EEG signals, exceeds a preset third conductivity threshold (e.g., 0.2 microseconds), the expression category obtained from facial expression image detection is a furrowed brow or downturned corners of the mouth, and the user's voice command is "too painful." A massage comfort score S∈[0,1] is obtained by weighted fusion of these multiple signals. When the massage comfort score S is less than a preset comfort threshold (e.g., 0.5), the user's current state is determined to be unpleasant (i.e., uncomfortable). The massage comfort score is calculated every certain time interval (e.g., 15 seconds). If the massage comfort score is less than the comfort threshold twice consecutively, the process automatically returns to step 101 to re-estimate the acupoint location and massage movement trajectory. Finally, the user controls the massage to stop by issuing a command. Furthermore, after each massage, the user's massage record can be automatically saved with the user's permission. This allows for real-time adjustments to subsequent massage paths and parameters based on the user's massage record, thereby further enhancing the user's personalized massage effect and user experience.

[0114] In practical applications, machine learning models such as convolutional neural networks can be used to assess massage comfort. This allows for the use of multimodal signals, including facial expressions, EEG signals, TENS signals, heart rate signals, and skin temperature, to jointly score comfort and ensure more reliable results. Furthermore, through a slow, closed-loop comfort assessment, the massage rhythm and duration for a specific user can be progressively optimized based on these multimodal signals, thereby enhancing the personalized massage experience.

[0115] It should be noted that although the aforementioned multimodal signals only involve facial expression images, electroencephalogram (EEG) signals, electrodermal signals, heart rate signals, and skin temperature, in specific applications, other indicators such as pulse signals, respiratory signals, electromyographic signals, and blood oxygen saturation can be added as needed in the adjustment of massage parameters and comfort scoring to construct a more comprehensive massage parameter adjustment strategy and comfort assessment model. This disclosure does not impose any limitations on this aspect.

[0116] Figure 2An exemplary implementation process of a massage robot performing massage operations is shown. See also Figure 2 The process of a massage robot performing a massage operation may include:

[0117] The user sits on the massage chair of the massage robot and begins coarse adjustments. The headrest and backrest are automatically adjusted, the user's posture is judged based on the pressure difference between the left and right sides of the backrest, and the posture is recognized by multiple sensors (i.e., 3D vision sensors, pressure sensor arrays, etc.). The posture is judged by detecting the deviation between the user's body center of gravity and the pressure center of the massage robot. If the posture is inconsistent, the recognition is re-performed and returned to the multi-sensor posture recognition. If the posture is consistent, acupoint mapping is performed, and the initial path and mode settings are completed.

[0118] In the fine-tuning stage, the robotic arm is moved to the tolerance range (i.e., the local area of ​​the first position). The position of each acupoint on the massage movement trajectory is adjusted through refined contact and pressure distribution analysis (i.e., the second position of each acupoint in the massage path is determined). The massage operation is performed according to the adjusted acupoint position.

[0119] During the massage, the system monitors the user's mood (i.e., the comfort level mentioned earlier). If discomfort is detected, the intensity is reduced; if pleasure is detected, the current intensity is maintained. The system also monitors for multiple consecutive instances of unpleasantness. If so, the system automatically reverts to the coarse adjustment stage; otherwise, the current intensity is maintained until the user selects the end.

[0120] As can be seen from the above, the embodiments of this disclosure propose a hierarchical control framework of "coarse adjustment - fine adjustment - real-time feedback". In the coarse adjustment stage, the user's human body point cloud data and virtual skeleton model are used to complete the identification of key skeletal points, the initial coordinate positioning of acupoints, and the preset of massage paths. In the fine adjustment stage, pressure distribution and light touch scanning are used to further find "refined massage points", effectively improving the accuracy of acupoint positioning, significantly improving the massage effect and enhancing the user experience. In addition, a closed-loop correction mechanism of physiological signals and user interaction feedback is introduced during the massage process, thereby ensuring the accuracy, personalization, and comfort of the massage process.

[0121] This disclosure first proposes the concept of "refined massage points" (i.e., the second location of acupoints and their local areas) and their dynamic positioning. Through the micro-force scanning of the robotic arm and synchronous analysis of physiological signals such as EEG, GSR, and heart rate, the optimal massage point within the tolerance zone is automatically selected and tracked when the user's posture changes. This breaks through the limitation of existing massage devices that can only perform actions at preset acupoints, and realizes personalized and precise massage by the massage robot.

[0122] Meanwhile, this disclosure also proposes an adaptive optimization mechanism for the up-and-down movement of the headrest and the angle of the backrest. Through pressure distribution data and posture analysis, it automatically finds the best support points for the neck and back, ensuring the user's body symmetry and comfort during long-term massage.

[0123] Furthermore, this embodiment constructs a multi-layered closed-loop control system, coupling three layers: a millisecond-level execution force closed loop, a second-level physiological signal closed loop, and a user interaction closed loop. This enables real-time correction of massage force output, dynamic maintenance of user comfort, and immediate response to individualized commands. This control mode significantly improves the system's adaptability and response speed to user needs.

[0124] Through the embodiments of this disclosure, users can not only make real-time corrections to massage parameters through voice commands (such as "lighter" or "turn to the left"), but also make passive feedback corrections to massage parameters through facial expression recognition, electroencephalogram (EEG) signals, electrodermal signals, heart rate signals, and skin temperature. This allows the system to automatically determine comfort levels and adjust massage parameters without the user's conscious awareness.

[0125] In this embodiment, coarse adjustment provides a macro framework, fine adjustment ensures micro precision, voice and facial expression interaction provide subjective corrections for users, and the headrest and backrest modules further enhance overall comfort. It can continuously converge to the optimal state with millimeter-level precision and second-level feedback, truly realizing an intelligent and personalized full-body massage solution.

[0126] Figure 3 A schematic diagram of the structure of the massage robot control device provided in an embodiment of this disclosure is shown. See also: Figure 3 The massage robot control device of this disclosure embodiment may include:

[0127] The coarse adjustment module 301 is used to determine the massage path. The determination of the massage path includes: acquiring the user's human body point cloud data, determining the massage path based on the user's human body point cloud data to obtain the user's massage path data. The massage path data includes the massage movement trajectory and the first position, first tolerance range and massage intensity range of each acupoint on the massage movement trajectory.

[0128] The fine-tuning module 302 is used to determine the second position of each acupoint on the massage movement trajectory. Determining the second position of each acupoint on the massage movement trajectory includes: controlling the robotic arm to perform multi-point light touches in a local area of ​​the first position of the acupoint and obtaining the pressure value of each light touch point; determining the second position based on the pressure value of each light touch point; the local area of ​​the first position is determined based on the first position and the first tolerance range; and the second position is the position of the light touch point with the smallest change in pressure value in the neighborhood among all the light touch points.

[0129] The massage operation module 303 is used to perform massage operations, which include controlling the robotic arm to perform massage operations according to the massage movement trajectory and the second position of each acupoint on the massage movement trajectory, the preset second tolerance range and the massage intensity range, wherein the second tolerance range is smaller than the first tolerance range.

[0130] Furthermore, the massage robot control device 300 of this embodiment may also include: a backrest adjustment module 304, used to detect the pressure difference between the left and right sides of the backrest when the user's back is in contact with the backrest of the massage robot in a sitting posture, and adjust the tilt of the backrest of the massage robot according to the pressure difference between the left and right sides of the backrest so that the pressure difference between the left and right sides of the backrest is less than or equal to a preset first pressure difference threshold.

[0131] Furthermore, the coarse adjustment module 301 can be specifically used to: construct a virtual skeleton model of the user using the user's human body point cloud data; perform acupoint mapping based on the user's virtual skeleton model and a pre-configured acupoint feature database to determine the user's massage path and generate massage path data.

[0132] Furthermore, the coarse adjustment module 301 can also be used to: detect the deviation between the user's body center of gravity and the pressure center of gravity of the massage robot in a sitting posture and adjust the virtual skeleton model according to the deviation so that the virtual skeleton is consistent with the user's body posture.

[0133] Furthermore, the massage robot control device 300 of this embodiment may further include: a massage parameter adjustment module 305, used to acquire the user's electroencephalogram (EEG) signal, electrodermal signal, heart rate signal, facial expression image, skin temperature and / or user instructions during the massage operation, and adjust the massage parameters of the massage operation according to the user's EEG signal, electrodermal signal, facial expression image, skin temperature and / or operation instructions. The massage parameters include the second position of the acupoint, the massage dwell time and / or the massage intensity.

[0134] Furthermore, the massage robot control device 300 of this embodiment may further include: a comfort assessment module 306, used to acquire the user's electroencephalogram (EEG), electrodermal (EDS), heart rate, skin temperature and / or facial expression images during the massage operation, and periodically assess the user's massage comfort based on the user's EEG, EDS, heart rate, skin temperature and / or facial expression images during the massage; if the user's massage comfort is lower than a preset comfort threshold after N consecutive assessments, the coarse adjustment module is notified to re-execute the step of determining the massage path, where N is an integer greater than 1 and a preset value is taken.

[0135] In practical applications, the control device 300 of the massage robot can be implemented as software, hardware, or a combination of both. For example, the control device 300 of the massage robot can be implemented as, but is not limited to, the control unit 401 of the massage robot 400 described below, or software running on the control unit 401.

[0136] Figure 4 A schematic diagram of the structure of the massage robot provided in an embodiment of this disclosure is shown. See also... Figure 4The massage robot may include: a control unit 401, a 3D vision sensor 402, a robotic arm 403, and a robotic hand 404. The robotic hand 404 is equipped with a second pressure sensor 405. The 3D vision sensor 402, the robotic arm 403, and the second pressure sensor 405 are respectively connected to the control unit 401, and the robotic hand 404 is connected to the robotic arm 403. The control unit 401 may include one or more processors and a memory for storing programs. The programs include instructions, which, when executed by the processor, cause the processor to perform the aforementioned control method of the massage robot.

[0137] The 3D vision sensor 402 is used to collect point cloud data of the user's human body.

[0138] The second pressure sensor 405 is used to detect the massage intensity of the massage operation, and can also be used to detect the pressure value of each light touch point in the local area of ​​the first position of the acupoint. In specific applications, the second pressure sensor 405 can be set in the area where the fingers of the robotic hand 404 contact the human body and apply massage pressure. For example, if the robotic hand 404 is a highly bionic robotic hand 404 finger that closely resembles the shape of a human hand, the second pressure sensor 405 can be placed on the fingertips of each finger of the robotic hand 404, the joints, and other parts that contact the human body and apply massage force during massage.

[0139] In practical applications, massage robots are realized as smart massage chairs, etc.

[0140] Furthermore, the massage robot may also include a first pressure sensor array 406, which can be used to detect the pressure generated by the contact and interaction between the human body and the massage robot when the human body is supported by the massage robot. Specifically, the first pressure sensor array 406 is arranged in the backrest, seat cushion, neck support, headrest, and other parts of the massage robot that support the human body.

[0141] Furthermore, the massage robot of this embodiment may also include one or more of the following sensors, which are respectively connected to the control unit 401: an electroencephalogram (EEG) sensor 407, a skin conductance sensor 408, a heart rate sensor 409, a skin temperature sensor 410, and a human-computer interaction component 411. The EEG sensor 407 is used to collect the user's brainwave signals. The skin conductance sensor 408 is used to collect the user's skin conductance signals, i.e., the conductivity or resistance value of the skin. The heart rate sensor 409 is used to collect the user's heart rate signals. Specifically, the heart rate signal may include, but is not limited to, instantaneous heart rate, average heart rate, etc. The skin temperature sensor 410 can be used to detect the user's skin temperature, especially the skin temperature of the massage operation area. The human-computer interaction component 411 may include, but is not limited to, touch buttons, a microphone, an eye-tracking component, a gesture recognition component, etc., and is used to receive user commands to adjust massage parameters according to the user commands. This embodiment does not limit the type of human-computer interaction component.

[0142] Depending on the needs, the massage robot 400 may also include other sensors such as a breathing sensor and an electromyography sensor, which are not limited in this embodiment.

[0143] In addition, embodiments of this disclosure also provide a computer-readable storage medium having a computer program stored thereon, the program including instructions that, when executed by one or more processors of a computing device, perform the steps of the aforementioned control method for the massage robot.

[0144] The aforementioned programs (also known as software, software applications, or code) include the machine instructions of a programmable processor and can be implemented using object-oriented programming languages, assembly language, or machine language.

[0145] With the development of time and technology, the meaning of "medium" has become increasingly broad. The dissemination of computer programs is no longer limited to tangible media; they can also be downloaded directly from the network. Any combination of one or more computer-readable storage media can be used. Computer-readable storage media can be, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or apparatus.

[0146] The technical solutions provided in this disclosure have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this disclosure. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this disclosure. Furthermore, those skilled in the art will recognize that, based on the ideas of this disclosure, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this disclosure.

[0147] The above are merely preferred embodiments of this disclosure and are not intended to limit this disclosure. Any modifications or equivalent substitutions made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A control method of a massage robot, characterized by, The method comprises: determining a massage path, the determining a massage path comprising: obtaining point cloud data of a human body of a user, determining a massage path based on the point cloud data of the human body of the user to obtain massage path data of the user, the massage path data comprising a massage movement track and first positions, first tolerance intervals and massage intensity intervals of acupoints on the massage movement track; determining second positions of the acupoints on the massage movement track, the determining second positions of the acupoints on the massage movement track comprising: controlling a mechanical hand to perform multi-point light touch in a local area of the first position of the acupoint and obtaining pressure values of the light touch points, determining the second positions according to the pressure values of the light touch points, the local area of the first position being determined according to the first position and the first tolerance interval, the second position being a position of a light touch point with the smallest change in pressure value in a neighborhood among all the light touch points; performing a massage operation, the performing a massage operation comprising: controlling the mechanical hand to perform a massage operation according to the massage movement track and the second positions, second tolerance intervals and the massage intensity intervals of the acupoints on the massage movement track, the second tolerance intervals being smaller than the first tolerance intervals.

2. The method according to claim 1, characterized in that, Before the determining a massage path based on the point cloud data of the human body of the user to obtain massage path data of the user, the method further comprises: detecting a pressure difference between left and right sides of a backrest of the massage robot when the back of the user in a riding posture contacts the backrest, and adjusting an inclination of the backrest of the massage robot according to the pressure difference between the left and right sides of the backrest to make the pressure difference between the left and right sides of the backrest less than or equal to a preset first pressure difference threshold.

3. The method according to claim 1, characterized in that, The determining a massage path based on the point cloud data of the human body of the user to obtain massage path data of the user comprises: constructing a virtual skeleton model of the user by using the point cloud data of the human body of the user; performing acupoint mapping based on the virtual skeleton model of the user and a preconfigured acupoint feature database to determine a massage path of the user and generate the massage path data.

4. The method according to claim 3, characterized in that, The determining a massage path based on the point cloud data of the human body of the user to obtain massage path data of the user further comprises: detecting a deviation between a body gravity center of the user in a riding posture and a pressure centroid of the massage robot and adjusting the virtual skeleton model according to the deviation to make the virtual skeleton model consistent with the body posture of the user.

5. The method of claim 1, characterized in that, The performing a massage operation comprises: when a deviation between a massage intensity of the massage operation and a massage intensity interval of a corresponding acupoint exceeds a preset massage intensity deviation range, adjusting a position of a mechanical arm until the deviation between the massage intensity of the massage operation and the massage intensity interval of the corresponding acupoint is less than a preset deviation threshold.

6. The method of claim 1, characterized in that, The performing a massage operation further comprises: acquiring electroencephalogram signals, electrodermal signals, heart rate signals, expression images, skin temperatures and / or user instructions of the user during the execution of the massage operation, and adjusting massage parameters of the massage operation according to the electroencephalogram signals, electrodermal signals, expression images, skin temperatures and / or operation instructions of the user, the massage parameters comprising the second positions of the acupoints, massage residence time lengths and / or massage intensities.

7. The method of claim 1, characterized in that, The performing a massage operation further comprises: During the massage operation execution, an electroencephalogram signal, a galvanic skin response signal, a heart rate signal, a skin temperature, and / or an expression image of the user are acquired, and the user's massage comfort level is evaluated in time according to the electroencephalogram signal, the galvanic skin response signal, the heart rate signal, the skin temperature, and / or the expression image of the user during the massage operation. If the massage comfort level of the user is lower than a preset comfort threshold for N consecutive times, the step of determining the massage path is re-executed, N is an integer greater than 1, and is a preset value.

8. A control device of a massage robot characterized by comprising: Comprise: A coarse adjustment module for determining a massage path, the determination of the massage path comprising: acquiring point cloud data of a human body of a user, determining a massage path based on the point cloud data of the human body of the user to obtain massage path data of the user, the massage path data comprising a massage moving track and a first position, a first tolerance interval, and a massage intensity interval of each acupoint on the massage moving track; A fine adjustment module for determining a second position of each acupoint on the massage moving track, the determination of the second position of each acupoint on the massage moving track comprising: controlling a mechanical hand to perform multi-point light touch in a local area of the first position of the acupoint and acquiring a pressure value of each light touch point, and determining the second position according to the pressure value of each light touch point, the local area of the first position being determined according to the first position and the first tolerance interval, and the second position being the position of the light touch point with the smallest change in pressure value in the neighborhood among all the light touch points; A massage operation module for executing a massage operation, the execution of the massage operation comprising: controlling a mechanical hand to execute a massage operation according to the massage moving track, the second position of each acupoint on the massage moving track, a second tolerance interval preset in advance, and the massage intensity interval, the second tolerance interval being smaller than the first tolerance interval.

9. A massaging robot characterized by comprising: The massage robot comprises a control unit, a 3D vision sensor, a mechanical arm, and a mechanical hand, the mechanical hand being provided with a second pressure sensor, the 3D vision sensor, the mechanical arm, and the second pressure sensor being connected to the control unit respectively, and the mechanical hand being connected to the mechanical arm; wherein the control unit comprises one or more processors and a memory storing programs, the programs comprising instructions which, when executed by the processors, cause the processors to execute the method of any one of claims 1-7.

10. A computer readable storage medium storing programs, the programs comprising instructions which, when executed by one or more processors of a computing device, cause the computing device to execute the method of any one of claims 1-7.

Images