Intelligent knocking massage path planning method

Abstract

The invention discloses an intelligent knocking massage path planning method which comprises the following steps: scanning a massage part of a user, and constructing a muscle acupoint topological map based on a preset traditional Chinese medicine acupoint diagram; performing muscle state monitoring according to the muscle acupoint topological map; processing the collected original data to extract multi-domain features; marking an obstacle avoidance area; knock path planning is carried out by combining the muscle academic topological map and the obstacle avoidance area; knocking massage is started according to the planned knocking path, and dynamic adjustment is conducted in the knocking massage process; the traditional Chinese medicine acupoint theory and the modern data monitoring technology are fused, the acupoint pertinence of traditional massage and the accuracy of the intelligent technology are considered, the limitation of a traditional massage fixed mode is avoided, the massage effect is improved, the muscle state and muscle inflammation are recognized through multi-domain features, the obstacle avoidance area is marked, and the massage effect is improved. And the massage scheme is adjusted based on the part form and the muscle state of the user, and the massage path is adjusted by circularly monitoring the muscle state in real time, so that the massage safety and effectiveness are ensured.

Description

Technical Field

[0001] This invention relates to the field of massage techniques, and more particularly to an intelligent tapping massage path planning method. Background Technology

[0002] As people's pace of life accelerates and their health awareness increases, massage, as an important way to relieve muscle fatigue and improve blood circulation, has seen its technical requirements upgrade from basic relaxation to personalized, precise, and safe services.

[0003] Traditional manual massage relies on the experience and judgment of professional technicians. The massage intensity and acupoint location vary depending on the experience of the physician, making it impossible to form a unified service standard. This can easily lead to significant differences in the massage effect of different technicians on the same area. Physicians can only judge contraindicated areas such as bruises and scars based on the user's verbal feedback or visual observation. If the user's description is unclear or the contraindicated area is hidden, it can easily cause secondary stimulation.

[0004] Ordinary mechanical massage, such as massage chairs and handheld massagers, achieves automation through preset fixed programs, without taking into account differences in user height, body type, etc., resulting in poor contact between the massage head and the skin and large deviations in acupoint positioning.

[0005] In conclusion, existing massage techniques cannot simultaneously meet the core requirements of precise shape adaptation, precise acupoint recognition, and intelligence. They cannot adjust massage plans according to the differences in body shape among different users to achieve personalized, safe, and effective massage. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide an intelligent tapping massage path planning method to solve the problems mentioned in the background section. To achieve the above objective, the present invention employs the following technical solution: This intelligent tapping massage path planning method includes the following steps: Step S1: Scan the user's massage area and generate a 3D model; Step S2: Based on the preset TCM acupoint map, mark the key points of acupoints in the 3D model and construct a muscle acupoint topology map; Step S3: Monitor muscle status based on the muscle acupoint topology map, including collecting raw data of the muscles at the tapping site during tapping. The raw data includes status data, vibration data, and temperature data. Step S4: Preprocess the collected raw data and extract multi-domain features, including time-domain features, frequency-domain features, and time-frequency-domain features. Step S5: Identify muscle status and muscle inflammation based on multi-domain features, and mark obstacle avoidance areas; Step S6: Combine the muscle acupoint topology map and obstacle avoidance area to plan the tapping path; Step S7: Begin tapping massage according to the planned tapping path, and continuously cycle through steps S3-S6 during the tapping massage process for dynamic adjustments.

[0007] Optionally, step S1 includes: scanning the contour of the massage area along a preset path using an infrared thermal imager and a millimeter-wave radar, recording the temperature value of each pixel on the path, and generating a 3D model by combining the current scanning position and scanning distance.

[0008] Optionally, step S3 includes: Changes in muscle resistance were collected using millimeter-wave radar. Data on muscle deformation during compression is collected using a pressure sensor; The vibration signals of the muscles during the massage are collected in real time using a triaxial accelerometer; By using an infrared thermal imager to collect the surface temperature distribution of the massaged area, potential high-temperature areas can be identified.

[0009] Optionally, step S4 includes: The collected raw data is filtered to remove environmental interference noise, and the data is segmented in time and space to form segmented data. Based on each segment of data, time-domain features, frequency-domain features, and time-frequency-domain features are extracted. The time-domain features include the mean and variance of muscle deformation, the frequency-domain features include the dominant frequency of muscle vibration, and the time-frequency-domain features include the instantaneous changes in muscle state.

[0010] Optionally, step S5 includes: Based on time domain features, frequency domain features, and time-frequency domain features, muscle status labels are output in real time. If bruise areas, surgical scar areas, and inflammatory areas are identified, they are marked as obstacle avoidance areas.

[0011] Optionally, identifying the inflammatory area includes: Extract candidate regions from segmented data whose average regional temperature is greater than a preset temperature; Calculate the temperature difference between the candidate region and its adjacent regions to exclude areas affected by high environmental temperatures. The remaining candidate regions were verified using pressure sensors and millimeter-wave radar, including indicators of muscle stiffness, resistance change, and tremor response.

[0012] Optionally, step S6 includes the following steps: Convert the muscle degree topology map into a raster map; Based on the obstacle avoidance area, each grid cell is labeled with its passage status and priority weight to form an attribute grid map; Set the start and end points; Cluster analysis is performed on the core target area grids in the attribute grid map to group the core target areas of multiple grids at a preset distance into a single target cluster; Calculate the center grid of each target cluster and sort them according to the Manhattan distance between the center grid and the starting point to form the target cluster access order; Generate a path sequence based on the access order of the target cluster.

[0013] Optionally, generating a path sequence based on the target cluster access order includes: Generate a path from the starting point to the center grid of the first target cluster; The tapping coverage grid within each target cluster is set according to the properties of the massage head. Starting from the last covered raster of the first target cluster, generate a path to the center raster of the next target cluster; The path passes through all target clusters in sequence until all target clusters are covered, forming a path sequence.

[0014] Optionally, adjusting the tapping parameters includes: setting the movable direction of the massage head based on a grid map; Set the step size for each movement; Set a coverage threshold for each target raster.

[0015] Optionally, the data spatiotemporal segmentation includes segmentation by time and spatial dimensions. The time-dimension segmentation includes: determining the segmentation time window and step size; extracting data according to the segmentation time window; and adding attribute labels to each data segment. The segmentation by spatial dimension includes: dividing fixed spatial regions according to body parts; and calculating the temperature label for each spatial region.

[0016] Compared to existing technologies, the advantages of this invention are that by integrating traditional Chinese medicine acupoint theory with modern data monitoring technology, it combines the acupoint targeting of traditional massage with the precision of intelligent technology, avoiding the limitations of fixed patterns in traditional massage. It meets the core requirements of precise form adaptation, accurate acupoint recognition, and intelligence, thus improving massage effectiveness. Through multi-domain feature identification of muscle state and inflammation, it marks obstacle avoidance areas and adjusts the massage plan based on the user's actual body shape and muscle state. By real-time cyclic monitoring of muscle state and adjusting the massage path, it ensures the safety and effectiveness of massage, avoids discomfort caused by changes in muscle state, and enhances the user experience. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the intelligent tapping massage path planning method of the present invention; Figure 2This is a schematic diagram of the process for monitoring muscle status based on a muscle acupoint topology map according to the present invention. Figure 3 This is a schematic diagram of the process for planning a tapping path by combining a muscle acupoint topology map and an obstacle avoidance area according to the present invention. Detailed Implementation

[0018] To facilitate understanding of this application, a more detailed description of the application is provided below with reference to the accompanying drawings and specific embodiments; preferred embodiments of the application are shown in the drawings; however, the application may be implemented in many different forms and is not limited to the embodiments described in this specification; rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of this application.

[0019] It should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances. In the embodiments shown in the accompanying drawings, directional indications (such as up, down, left, right, front, and back) are used to explain the structure and movement of various components and are not absolute but relative. These descriptions are appropriate when these components are in the positions shown in the drawings. If the descriptions of the positions of these components change, these directional indications also change accordingly.

[0020] It should also be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence; it should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in a sequence other than those illustrated or described herein.

[0021] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit this application.

[0022] like Figure 1 As shown, one embodiment of the present invention is a smart tapping massage path planning method, which includes the following steps: Step S1: Scan the user's massage area, such as the shoulders and back, to obtain the contour and shape data of the area and build a three-dimensional model of the area. Step S2: Combine the preset standard atlas of Chinese acupoints, accurately mark the key points of acupoints on the 3D model generated in step S1, establish a topological map of the relationship between muscle morphology and acupoint location, and determine the spatial positional relationship between each acupoint and the surrounding muscles. Step S3: Monitor muscle status based on the muscle acupoint topology map, including collecting raw data of the muscles at the tapping site during tapping. The raw data includes status data, vibration data, and temperature data. Specifically, based on the topology map of S2, raw data of the muscles at the tapping site are collected during the tapping test, including state data reflecting the basic condition of the muscles, vibration data reflecting the vibration feedback of the muscles, and temperature data of the tapping site. Step S4: Preprocess the collected raw data and extract multi-domain features, including time-domain features, frequency-domain features, and time-frequency-domain features. Specifically, the raw data collected by S3 is processed and the time domain features, i.e. the data change patterns over time, the frequency domain features, i.e. the frequency distribution characteristics of the data, and the time-frequency domain features, i.e. the joint change characteristics of the data in the time and frequency dimensions, are extracted. Step S5: Identify muscle status and muscle inflammation based on multi-domain features, and mark obstacle avoidance areas; Specifically, using the multi-domain features extracted by S4, the current state of the muscle is determined, such as tension, relaxation, and whether there is inflammation. If areas that are not suitable for tapping, such as bruises, surgical scars, or inflammation, are identified, they are marked as obstacle avoidance areas. Step S6: Combine the muscle acupoint topology map and obstacle avoidance area to plan the tapping path; Specifically, by integrating the muscle and acupoint topology map of S2 with the obstacle avoidance area marked by S5, the tapping movement path of the massage head is planned so that the path covers the target acupoint and avoids the obstacle avoidance area. Step S7: Begin tapping massage according to the planned tapping path, and continuously cycle through steps S3-S6 during the tapping massage process to make dynamic adjustments, ensuring the effectiveness and comfort of the massage process, and avoiding massage intensity or discomfort due to changes in muscle state, such as muscles gradually relaxing and local temperature rising during the massage.

[0023] This application integrates traditional Chinese medicine acupoint theory with modern data monitoring technology, taking into account the acupoint targeting of traditional massage and the precision of intelligent technology. It avoids the limitations of the fixed pattern of traditional massage, improves the massage effect, identifies muscle state and muscle inflammation through multi-domain feature recognition, marks obstacle avoidance areas, and adjusts the massage plan based on the user's actual body part shape and muscle state. By monitoring muscle state in real time and adjusting the massage path, it ensures the safety and effectiveness of massage, avoids discomfort caused by changes in muscle state, and improves the user experience.

[0024] In one embodiment, step S1 includes: scanning the contour of the massage area along a preset path using an infrared thermal imager and a millimeter-wave radar, recording the temperature value of each pixel on the path, and generating a 3D model by combining the current scanning position and scanning distance.

[0025] Specifically, an infrared thermal imager and a millimeter-wave radar work together to scan the contour of the massage area along a preset path, such as a path that conforms to the curve of the massage area, such as the neck, shoulders, and back. The infrared thermal imager records the temperature value of each pixel on the scanning path to obtain surface temperature distribution data of the area. The millimeter-wave radar measures the current scanning position and scanning distance. The contour, temperature, position, and distance data collected by the two devices are fused and processed to generate a 3D model of the massage area that includes temperature information.

[0026] In one embodiment, such as Figure 2 As shown, step S3 includes: Step S31: Collect changes in muscle resistance using millimeter-wave radar, and determine changes in muscle density and tension by using the reflected signal after the radar waves penetrate the surface tissue. Step S32: Collect muscle deformation data during pressing using a pressure sensor, record the correspondence between pressing force and deformation degree, and reflect muscle elasticity; Step S33: Collect muscle tremor signals in real time during the massage using a triaxial accelerometer, and capture feedback data such as vibration frequency and amplitude after the muscles are tapped. Step S34: Collect the surface temperature distribution of the massage area using an infrared thermal imager to identify potential high-temperature areas; collect the surface temperature distribution of the massage area and compare it with the temperature data of the initial 3D model to identify potential high-temperature areas that occur during the massage process.

[0027] By using multiple sensors in synergy, we can achieve multi-dimensional and three-dimensional monitoring of muscle status, covering the physical characteristics, dynamic feedback, and metabolic status of muscles. This avoids the limitations of data from a single sensor, and data from different sensors can be cross-validated, reducing the risk of misjudgment from a single data source.

[0028] In one embodiment, step S4 includes: The collected raw data is filtered to remove environmental interference and noise, ensuring the authenticity and validity of the data, and the data is segmented in time and space to form segmented data. Based on each segment of data, time-domain features, frequency-domain features, and time-frequency-domain features are extracted. The time-domain features include the mean and variance of muscle deformation, the frequency-domain features include the dominant frequency of muscle vibration, and the time-frequency-domain features include the instantaneous changes in muscle state.

[0029] Understandably, the time-domain features, namely the mean of muscle deformation, reflect the average degree of deformation, while the variance reflects the range of deformation fluctuations, embodying the statistical characteristics of the data over time; the frequency-domain features, namely the dominant frequency of muscle vibration, reflect the main frequency components of the tremor signal, embodying the frequency distribution pattern of the data; and the time-frequency domain features, namely the instantaneous changes in muscle state, such as the synchronous changes in temperature and vibration frequency at a certain moment, embody the joint characteristics of the data in the time and frequency dimensions, transforming the raw data into feature indicators that can be used for state recognition.

[0030] In one embodiment, step S5 includes: Based on time domain features, frequency domain features, and time-frequency domain features, muscle status labels are output in real time. If bruise areas, surgical scar areas, and inflammatory areas are identified, they are marked as obstacle avoidance areas.

[0031] Specifically, based on the time-domain, frequency-domain, and time-frequency-domain features extracted by S4, the current state of the muscle is judged in real time through a preset recognition model (such as a machine learning model), and the corresponding state label is output, such as muscle tension, muscle relaxation, suspected inflammation, etc. If three types of unsuitable areas for tapping are identified through features: bruised areas, such as abnormal deformation data or slow vibration decay; surgical scar areas, such as muscle resistance mutation or irregular shape; and inflamed areas, such as elevated temperature or abnormal resistance, these areas are marked as obstacle avoidance areas, and the massage head is prohibited from entering these obstacle avoidance areas.

[0032] By comprehensively judging multi-domain features, misjudgments caused by single features are avoided. For example, high temperature alone may be due to environmental factors, but combining resistance features can more accurately judge inflammation. By marking obstacle avoidance areas in advance, risks can be avoided, and the problems of traditional massage fixation mode can be solved. It is especially suitable for users with local injuries.

[0033] In one embodiment, identifying the inflammatory area includes: Candidate regions with an average regional temperature greater than a preset temperature are extracted from segmented data as inflammation candidate regions. Calculate the temperature difference between the candidate region and the adjacent region, and exclude regions affected by high ambient temperature. If the temperature difference is less than a preset threshold, it is determined to be a region affected by high ambient temperature and is excluded, such as the influence of external heat sources.

[0034] The remaining candidate regions were verified using pressure sensors and millimeter-wave radar, including muscle stiffness indicators, resistance change indicators, and tremor response indicators. Specifically, the remaining candidate areas are further verified using pressure sensors and millimeter-wave radar. The pressure sensors detect muscle stiffness in the area, the millimeter-wave radar detects resistance changes using reflected signals, and the triaxial accelerometer monitors tremor response using vibration decay rate. If at least two of the three indicators match inflammatory characteristics, such as increased stiffness, increased resistance, and slow tremor decay, the area is confirmed as an inflammatory region. Identifying inflammatory regions avoids misjudging interference factors such as high environmental temperatures as inflammation, making obstacle avoidance area marking more reliable and preventing subsequent massage from stimulating the inflamed area, thus reducing the risk of secondary injury.

[0035] In one embodiment, such as Figure 3 As shown, step S6 includes the following steps: Step S61: Convert the muscle acupoint topology map into a grid map. Each grid corresponds to a fixed small area of ​​the actual massage site. Each tap of the massage head may contain multiple grids. Step S62: Mark the passage status and priority weight of each grid according to the obstacle avoidance area to form an attribute grid map. Mark the passage status of each grid in the grid map, such as: passable or impassable. At the same time, mark the priority weight according to the importance of acupoints and the degree of muscle tension to form an attribute grid map. Step S63: Set the start and end points. Based on the boundaries of the massage area, such as the upper edge of the back as the start point and the lower edge as the end point, or according to user needs, set the initial start point and final end point of the massage head. Step S64: Perform cluster analysis on the core target area grids in the attribute grid map, and group the core target areas of multiple grids with a preset distance into one target cluster; perform cluster analysis on the core target area grids with high priority weights in the attribute grid map, and group the core target area grids with a distance less than a preset value into one target cluster. Step S65: Calculate the center grid of each target cluster and sort them according to the Manhattan distance between the center grid and the starting point, i.e., the horizontal distance and the vertical distance, to form the target cluster access order; Step S66: Based on the target cluster access order, plan the complete path sequence of the massage head from the starting point to each target cluster and then to the end point, and generate the path sequence.

[0036] The rasterized map can flexibly adjust the grid size and clustering threshold according to different massage areas to adapt to the shape requirements of different human body parts. The clustering and priority labeling of the core target area grid ensures that high-importance areas are covered first and avoids missing key parts.

[0037] In one embodiment, generating a path sequence based on the target cluster access order includes: Generate a path from the starting point to the center grid of the first target cluster; Based on the properties of the massage head, such as the diameter of the massage head and the effective range, set the number and range of the coverage grids to be tapped within each target cluster; After completing the tapping of all the overlay grids in the first target cluster, use the last overlay grid of the target cluster as the new starting point to generate a path to the center grid of the next target cluster; The path passes through all target clusters in sequence until all target clusters are covered, ultimately forming a complete path sequence from the initial starting point to all target clusters and then to the endpoint.

[0038] In one embodiment, adjusting the tapping parameters includes: setting the movable direction of the massage head based on a grid map, such as the four directions of up, down, left, and right; and setting the movable direction of the massage head on each grid according to the adjacency relationship of the grid, so as to prevent the massage head from going beyond the grid map boundary or entering the obstacle avoidance area. Set the step size for each movement; based on the massage precision requirements and the size of the massage head, set the step size for each movement of the massage head, such as the step size being equal to the grid side length, to ensure that each movement covers a complete grid. Set a coverage threshold for each target grid, which is the number of times or the duration of action of the massage head on that grid. For example, tap the core acupoints five times and the ordinary area three times to ensure that the target grid is effectively covered.

[0039] The step size and coverage threshold can be flexibly adjusted according to user needs, such as reducing the step size or lowering the coverage threshold for sensitive users, to suit the tolerance of different users.

[0040] In one embodiment, data spatiotemporal segmentation includes segmentation by time and spatial dimensions. Segmenting by time dimension includes: determining the segment time window and step size; extracting data according to the segment time window; and adding attribute labels to each data segment. Determine the segmentation parameters: Based on the massage duration and data change frequency, set the segmentation time window and step size, such as 5 seconds for each window, 2 seconds for each step, or 3 seconds of window overlap. Data extraction: Extract the raw data collected in step S3 according to the time window to form a continuous time data segment; Add attribute tags: Add time attribute tags and corresponding massage action tags for each time data segment, such as 0-5 seconds tapping intensity level 3.

[0041] Segmentation by spatial dimension includes: dividing fixed spatial areas according to body parts; calculating the temperature label for each spatial area.

[0042] Fixed area division: According to the physiological structure of the human body, such as the back being divided into upper back, middle back, and lower back, and the neck being divided into upper neck, middle neck, and lower neck, fixed spatial areas are divided. Temperature label calculation: For each spatial region, calculate the average temperature of all pixels in that region and generate a spatial temperature label, such as the average temperature of the upper back being 37℃.

[0043] By extracting features from spatiotemporally segmented data, we can avoid the mixing of data from different times and locations, making the features more closely match the real state of the local muscles and further improving the accuracy of muscle state recognition.

[0044] It should be noted that the above-mentioned technical features can be combined with each other to form various embodiments not listed above, all of which are considered to be within the scope of the present invention specification; and, for those skilled in the art, improvements or modifications can be made based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for intelligent tapping massage path planning, characterized in that, Includes the following steps: Step S1: Scan the user's massage area and generate a 3D model; Step S2: Based on the preset TCM acupoint map, mark the key points of acupoints in the 3D model and construct a muscle acupoint topology map; Step S3: Monitor muscle status based on the muscle acupoint topology map, including collecting raw data of the muscles at the tapping site during tapping. The raw data includes status data, vibration data, and temperature data. Step S4: Preprocess the collected raw data and extract multi-domain features, including time-domain features, frequency-domain features, and time-frequency-domain features. Step S5: Identify muscle status and muscle inflammation based on multi-domain features, and mark obstacle avoidance areas; Step S6: Combine the muscle acupoint topology map and obstacle avoidance area to plan the tapping path; Step S7: Begin tapping massage according to the planned tapping path, and continuously cycle through steps S3-S6 during the tapping massage process for dynamic adjustments.

2. The intelligent tapping massage path planning method according to claim 1, characterized in that, Step S1 includes: scanning the contour of the massage area along a preset path using an infrared thermal imager and millimeter-wave radar, recording the temperature value of each pixel on the path, and generating a 3D model by combining the current scanning position and scanning distance.

3. The intelligent tapping massage path planning method according to claim 1, characterized in that, Step S3 includes: Changes in muscle resistance were collected using millimeter-wave radar. Data on muscle deformation during compression is collected using a pressure sensor; The vibration signals of the muscles during the massage are collected in real time using a triaxial accelerometer; By using an infrared thermal imager to collect the surface temperature distribution of the massaged area, potential high-temperature areas can be identified.

4. The intelligent tapping massage path planning method according to claim 1, characterized in that, Step S4 includes: The collected raw data is filtered to remove environmental interference noise, and the data is segmented in time and space to form segmented data. Based on each segment of data, time-domain features, frequency-domain features, and time-frequency-domain features are extracted. The time-domain features include the mean and variance of muscle deformation, the frequency-domain features include the dominant frequency of muscle vibration, and the time-frequency-domain features include the instantaneous changes in muscle state.

5. The intelligent tapping massage path planning method according to claim 1, characterized in that, Step S5 includes: Based on time domain features, frequency domain features, and time-frequency domain features, muscle status labels are output in real time. If bruise areas, surgical scar areas, and inflammatory areas are identified, they are marked as obstacle avoidance areas.

6. The intelligent tapping massage path planning method according to claim 5, characterized in that, The identification of inflammatory areas includes: Extract candidate regions from segmented data whose average regional temperature is greater than a preset temperature; Calculate the temperature difference between the candidate region and its adjacent regions to exclude areas affected by high environmental temperatures. The remaining candidate regions were verified using pressure sensors and millimeter-wave radar, including indicators of muscle stiffness, resistance change, and tremor response.

7. The intelligent tapping massage path planning method according to claim 1, characterized in that, Step S6 includes the following steps: Convert the muscle degree topology map into a raster map; Based on the obstacle avoidance area, each grid cell is labeled with its passage status and priority weight to form an attribute grid map; Set the start and end points; Cluster analysis is performed on the core target area grids in the attribute grid map to group the core target areas of multiple grids at a preset distance into a single target cluster; Calculate the center grid of each target cluster and sort them according to the Manhattan distance between the center grid and the starting point to form the target cluster access order; Generate a path sequence based on the access order of the target cluster.

8. The intelligent tapping massage path planning method according to claim 7, characterized in that, The path sequence generated based on the target cluster access order includes: Generate a path from the starting point to the center grid of the first target cluster; The tapping coverage grid within each target cluster is set according to the properties of the massage head. Starting from the last covered raster of the first target cluster, generate a path to the center raster of the next target cluster; The path passes through all target clusters in sequence until all target clusters are covered, forming a path sequence.

9. The intelligent tapping massage path planning method according to claim 1, characterized in that, The adjustment of the tapping parameters includes: setting the movable direction of the massage head based on a grid map; Set the step size for each movement; Set a coverage threshold for each target raster.

10. The intelligent tapping massage path planning method according to claim 4, characterized in that, The data spatiotemporal segmentation includes segmentation by time dimension and spatial dimension. The time-dimension segmentation includes: determining the segmentation time window and step size; extracting data according to the segmentation time window; and adding attribute labels to each data segment. The segmentation by spatial dimension includes: dividing fixed spatial regions according to body parts; and calculating the temperature label for each spatial region.

Images