KR20220121620A - Individual customized hybrid fitness coaching system and method - Google Patents

Abstract

A personalized hybrid fitness coaching system according to one embodiment of the present invention comprises: a data collection unit which collects body-related information of a user from a user terminal and a fitness center terminal; a body condition analysis unit which analyzes the physical condition and living habits of the user based on the body-related information; a fitness coaching information generation unit which generates an inference model by learning the body-related information of the user, and the analyzed body condition information and living habit information with artificial intelligence and generates fitness coaching information suitable for the user based on the generated inference model; and a fitness coaching information providing unit which provides customized fitness coaching content and programs including the generated fitness coaching information, current fitness status information and future fitness prediction information of the user. The system is capable of providing optimized fitness information based on physical information and living habit information.

Description

INDIVIDUAL CUSTOMIZED HYBRID FITNESS COACHING SYSTEM AND METHOD

The present invention relates to a fitness coaching system and method, and more particularly, personalized hybrid coaching combining online and offline that can provide optimal fitness coaching information by learning individual body information and lifestyle data with artificial intelligence. systems and methods.

In modern society, lifestyle-related diseases such as obesity and metabolic diseases are becoming a major social problem for each life cycle, regardless of age or gender. For the prevention and treatment of these diseases, WHO emphasizes the importance of regular fitness and nutrition.

However, modern people cannot be interested in fitness due to lack of fitness time due to busy daily life and lack of professional knowledge about fitness, and the number of times they are exposed to irregular meal times and instant food is increasing.

In addition, with the increase in Internet usage, nutrition information and nutrition counseling are provided by government agencies, schools, nutrition-related societies, medical institutions, or individual experts such as nutritionists and doctors through the Internet. Although programs are being developed, there are almost no programs that suggest customized diets in consideration of individual physical characteristics and fitness intensity.

In addition, the need to provide personalized fitness information is increasing due to the decrease in physical activity due to rapid social change and lack of fitness. The system does not exist.

Korean Patent Registration No. 10-2054134 (2019.12.04)

The present invention is to solve the above problems, and in detail, to provide a personalized hybrid fitness coaching system and method that can provide optimal fitness information based on an individual's updated body information and lifestyle information The purpose.

The problems of the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

Personalized hybrid fitness coaching system according to an embodiment of the present invention for solving the above problems includes a server connected to a user terminal and a fitness center terminal and a network, the server, from the user terminal and the fitness center terminal a data collection unit for collecting user's body-related information; a body condition analysis unit that analyzes the user's physical condition and lifestyle based on the body-related information; Fitness coaching information for generating an inference model by learning the user's body-related information, the analyzed body state information, and lifestyle information with artificial intelligence, and generating fitness coaching information suitable for the user based on the generated inference model generator; and a fitness coaching information providing unit that provides customized fitness coaching content and programs including the generated fitness coaching information, the user's current fitness status information and future fitness prediction information to the user terminal.

The fitness coaching information generator may generate the fitness coaching information by applying an artificial intelligence learning model to the body-related information updated based on a multi-armed bandit (MAB) algorithm.

The fitness coaching information generation unit may generate a fitness goal achievement prediction probability and fitness prediction result based on the updated body-related information and pre-built big data information for each gender, age, and body type.

The fitness coaching information generation unit may generate and combine a plurality of machine learning or deep learning-based predictive models through ensemble learning, and use them to generate the fitness goal achievement prediction probability and fitness prediction result.

The body-related information includes the user's photographed image information, an image analysis unit for extracting feature points from the photographed image information to analyze the user's body change information; and an image generation unit for generating image information of the user for the future within a fitness target period based on the analyzed body change information, wherein the fitness coaching information providing unit generates current image information of the user and the image generation unit and provide future image information to the user terminal.

The body-related information includes GPS sensor information, acceleration sensor information, and gyro sensor information received from the user terminal, and the body state analyzer is based on the GPS sensor information, the acceleration sensor information, and the gyro sensor information. The user's movement information may be calculated to analyze the user's physical condition and lifestyle.

The fitness coaching information may include aerobic exercise information, anaerobic exercise information, and FITT (frequency, intensity, time, form) information.

In addition, the personalized hybrid fitness coaching method according to an embodiment of the present invention comprises the steps of: (a) collecting the user's body-related information from the user terminal and the fitness center terminal; (b) analyzing the user's physical condition and lifestyle based on the body-related information; (c) generating an inference model by learning the user's body-related information, the analyzed body state information, and lifestyle information with artificial intelligence, and generating fitness coaching information suitable for the user based on the generated inference model step; And (d) the generated fitness coaching information, the user's current fitness status information, and may include customized coaching content and professional information including future fitness prediction information.

The step (c) may include generating the fitness coaching information by applying an artificial intelligence learning model to the body-related information updated based on a multi-armed bandit (MAB) algorithm.

The step (c) may include generating a fitness goal achievement prediction probability and fitness prediction result based on the updated body-related information and pre-built big data information for each gender, age, and body type.

The body-related information includes the user's photographed image information, and the step (b) includes extracting a feature point from the photographed image information to analyze the user's body change information, and the step (c) includes: , generating image information of the user for the future within a fitness target period based on the analyzed body change information; and providing the user's current image information and the generated future image information to the user terminal.

A personalized hybrid fitness coaching system and method according to an embodiment of the present invention can collect an individual's updated body information and lifestyle information, and provide optimal fitness information based thereon.

In addition, the personalized hybrid fitness coaching system and method according to an embodiment of the present invention can provide optimal fitness information on an online and offline basis by receiving body information and lifestyle information from a user terminal and a fitness center terminal.

In addition, the personalized hybrid fitness coaching system and method according to an embodiment of the present invention may predict an individual's fitness goal achievement probability and result based on big data and provide it to a user terminal.

The effect according to the present invention is not limited by the contents exemplified above, and more various effects are included in the present specification.

1 is a schematic configuration diagram showing a network to which a personalized hybrid fitness coaching system is applied according to an embodiment of the present invention.
2 is a detailed configuration diagram of a personalized hybrid fitness coaching system according to an embodiment of the present invention.
3 is an exemplary diagram of a fitness coaching scheduling program of a personalized hybrid fitness coaching system according to an embodiment of the present invention.
4 is a schematic diagram illustrating the concept of a personalized hybrid fitness coaching method according to an embodiment of the present invention.
5 is a flowchart illustrating a personalized hybrid fitness coaching method according to an embodiment of the present invention.
6 is a flowchart illustrating an image analysis and future image information provision method of a personalized hybrid fitness coaching method according to an embodiment of the present invention.

Advantages and features of the present invention and methods of achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments allow the disclosure of the present invention to be complete, and common knowledge in the technical field to which the present invention pertains It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

Hereinafter, a personalized hybrid fitness coaching system and method according to embodiments of the present invention will be described with reference to the drawings.

1 is a schematic configuration diagram showing a network to which a personalized hybrid fitness coaching system is applied according to an embodiment of the present invention.

Referring to FIG. 1 , the personalized hybrid fitness coaching system according to an embodiment of the present invention may be configured to include a server 100 , and the server 100 includes a user terminal 200 and a fitness center terminal 300 and It is connected to a wired or wireless network.

The user terminal 200 refers to a smart device such as a smart phone, a tablet, and a smart watch carried by a user.

The fitness center terminal 300 is a terminal owned by the manager of the fitness center and refers to devices such as a desktop computer, a laptop computer, a smartphone, a tablet, and a body composition measuring device.

The user terminal 200 may include a plurality of terminals 210 , 220 , and 230 owned by a plurality of users. Accordingly, the server 100 may be configured to collect body information of a plurality of users and provide coaching information to each of the plurality of users.

Although it is shown that the fitness center terminal 300 is configured as a single unit in FIG. 1 , like the user terminal 200 , it may be configured in plurality.

2 is a detailed configuration diagram of a personalized hybrid fitness coaching system according to an embodiment of the present invention.

Referring to FIG. 2 , the server 100 includes a data collection unit 110 , a body state analysis unit 120 , a fitness coaching information generation unit 130 , a fitness coaching information providing unit 140 , and an image analysis unit 150 . , the image generator 160 , the user body state DB 170 , the fitness model DB 175 , the fitness equipment DB 177 , and the attribute matching DB 180 may be configured.

The data collection unit 110 serves to collect the user's body-related information from the user terminal 200 and the fitness center terminal 300 .

The user's body-related information includes gender, height (cm), weight (kg), fat free mass (FFM), body fat mass (kg), BMI (body mass index, kg/m ² ), WHR (waist/hip girth ratio), BMR (basal metabolic rate, Kcal), user captured images, sensor information, and the like.

The user body-related information may be data input by a user or an administrator through a computer program or application provided by the server 100 of the present invention, or data measured by the user terminal 200 or the fitness center terminal 300 .

For example, the user terminal 200 may transmit the user's photographed image, GPS sensor information, acceleration sensor information, and gyro sensor information to the server 100 .

In addition, the fitness center terminal 300 may be a measuring device such as a dynamometer, a back muscle dynamometer, a left flexometer, a spirometer, a wireless heart rate monitor, and the like, and data of the measuring device may be transmitted to the server 100 .

The body condition analysis unit 120 analyzes the user's body condition and lifestyle based on the body-related information.

For example, the body condition analyzer 120 calculates the user's exercise amount based on the GPS sensor information, the acceleration sensor information, and the gyro sensor information collected from the user terminal 200 to analyze the physical condition and information on the amount of exercise according to time. Lifestyle can be analyzed based on

The fitness coaching information generation unit 130 generates an inference model by learning the user's body-related information, the analyzed body state information, and lifestyle information with artificial intelligence, and is suitable for the user based on the generated inference model. It plays a role in generating fitness information.

For example, the fitness coaching information generating unit 130 may initially generate a customized fitness exercise program based on the user's body measurement information collected from the fitness center terminal 300 . In this case, the fitness center terminal 300 may be a computer of the fitness center, a laptop computer, a measuring device such as a body composition measuring device, and the like.

In addition, the fitness coaching information generating unit 130 updates the user's body measurement information based on the user's body measurement information collected from the user terminal 200, and updates the body condition information and lifestyle information based on this information to the above. It is possible to generate fitness information suitable for the user.

The user's body condition DB 170 serves to store the user's body information collected by the data collection unit 110 and the user's body condition information and lifestyle information analyzed by the body condition analysis unit 120 .

The fitness model DB 175 serves to store fitness model information based on big data for each gender, age, and body type of the user. For example, the fitness model DB 175 may include information such as the fitness coaching scheduling program shown in FIG. 3 .

The fitness equipment DB 177 serves to include information on fitness equipment based on big data for each user's gender, age, and body type.

Attribute matching DB (177) is based on the user's body information, physical condition information, lifestyle information, fitness goal information, fitness goal period information and big data by gender, age, and body type, the optimal coaching corresponding to each attribute of the user It serves to store attribute matching information that matches information.

In addition, the attribute matching DB 177 may serve to store attribute matching information for matching optimal nutritional diet information and lifestyle information based on the user's body information, body state information, lifestyle information, fitness goal information, etc. have.

The fitness coaching information providing unit 140 may provide customized fitness coaching content and programs including the generated fitness coaching information, the user's current fitness status information, and future fitness prediction information to the user terminal 200 .

The fitness coaching information providing unit 140 may provide optimal fitness coaching information based on user constraints, and the user constraints include implicit constraints such as changes in physical condition or lifestyle, fitness goals, and preference strength. There may be explicit constraints such as preference changes.

The fitness coaching information providing unit 140 may provide a plurality of recommendation results for each ranking according to a preset criterion to the user terminal 200 using a ranking algorithm.

The fitness coaching information providing unit 140 may provide different coaching information for each service step.

For example, in the beginner hybrid customized fitness coaching program, big data-based lifestyle analysis information, fitness scheduling and strength program information according to individual physical conditions such as underweight, normal, overweight, and obesity according to BMI values, aerobic activity using a treadmill or stationary bicycle It is possible to provide sexual fitness scheduling and program information, and anaerobic fitness scheduling and program information that mobilizes isometric and isotonic contractions using one's own body weight.

In the intermediate hybrid customized fitness coaching program, anaerobic/anaerobic fitness training model, BMI (Body Mass Index) and %Fat complex-based BMI group-specific fat type or muscle type individual body characteristic classification information, F.I.T.T (frequency, intensity, time, present) principle based fitness program, interval training program by intensity considering incomplete rest between individual fitness performance period and repeated performance, various anaerobic fitness programs based on weight-bearing using a gym ball and bounty, small tool balance and strength fitness program information can be provided. have.

The advanced hybrid customized fitness coaching program provides a report for providing information on physical strength, body characteristics, and 1RM (One Repetition Maximum, maximum strength) measured by a professional trainer. Provides information, reinforcement and maintenance based on physical strength and physical characteristics, provides anaerobic/anaerobic fitness scheduling and programs based on physical fitness information, provides optimal fitness coaching information based on deep learning, and provides personalized muscle strength, hypertrophy, It can provide information on anaerobic fitness programs to improve power, endurance, agility and speed.

The physical condition analysis unit 120 may perform analysis by using information stored in the user's physical condition DB 170 , and the fitness coaching information generating unit 130 and the fitness coaching information providing unit 140 are the user's physical condition DB 170 , fitness model DB 175 , fitness equipment DB 177 , and information stored in attribute matching DB 180 may be used to generate and provide fitness coaching information.

The fitness coaching information providing unit 130 applies a learning model to the body-related information that is updated based on a multi-armed bandit (MAB) algorithm to provide customized fitness coaching information, nutritional diet information, and lifestyle information. can create

In addition, the fitness coaching information generating unit 130 may generate a fitness goal achievement prediction probability and fitness prediction result based on the updated body-related information and pre-built big data information for each gender, age, and body type.

In addition, the fitness coaching information generation unit 130 may generate and combine a plurality of machine learning or deep learning-based prediction models through ensemble learning, and generate a fitness goal achievement prediction probability and a fitness prediction result using this.

The present invention can perform more accurate final prediction than using a single classifier through ensemble learning. As a learning method, you can use the bagging method, which is a method of collecting samples by taking samples several times and learning each model to aggregate the results, and the boosting method, which is a weighting method that applies the results of one model to another model.

A fitness goal may be in a category such as 'diet' or 'muscle strengthening', and if it is a 'diet' goal, it may include information such as a specific target weight, and may include information about the entire period to achieve the target weight. can

For example, the fitness coaching information generator 130 may predict a long-term (eg, 52 weeks) fitness result based on big data. That is, the fitness coaching information generation unit 130 predicts whether the final goal is achieved by using the collected user's body information (BFM information, anthropometric information, information on whether or not the goal has been achieved according to the period) and can be expressed as a probability value. In addition, the fitness coaching information providing unit 140 may provide it to the user terminal 200 .

In this case, the fitness coaching information generating unit 130 may utilize big data by gender/age/body type, and an RNN, an LSTM model, and the like as a predictive model.

The body-related information collected by the data collection unit 110 from the user terminal 200 and the fitness center terminal 300 may include a photographed image of the user.

The image analysis unit 150 may analyze the user's body change information by extracting feature points from the photographed image information.

The image generator 160 serves to generate image information of the user for the future within a fitness target period based on the analyzed body change information.

The fitness coaching information providing unit 140 may provide the user's current image information and future image information generated by the image generating unit 160 to the user terminal 200 .

For example, the fitness coaching information providing unit 140 may generate a weekly, monthly, and yearly change image according to the progress of fitness and provide it to the user terminal 200, and determine whether the user has achieved the goal so far. Based on the continuous update of the future image based on the user's exercise, nutrition and lifestyle changes can be induced.

4 is a schematic diagram illustrating the concept of a personalized hybrid fitness coaching method according to an embodiment of the present invention.

The personalized hybrid fitness coaching method of the present invention is an online fitness by the personalized fitness content provided to the user terminal 200 by the server 100 of the present invention, and a personalized fitness program provided by the server 00 of the present invention. Includes offline fitness conducted in a fitness center by

In online fitness, when the user sets the ultimate goal of fitness with the user terminal 200, the user responds according to the customized fitness content provided by the server 100, and analyzes personal body information and lifestyle based on the user's reaction This includes providing customized fitness content again based on AI.

A user may visit a fitness center offline during online fitness and perform fitness according to a customized fitness program provided by the server 100 .

While performing the above online and offline processes, the server 100 may provide the results of personal fitness performance analysis online through the user terminal 200 at regular intervals, and when the target organization is reached, the final performance analysis results will be provided. can

5 is a flowchart illustrating a personalized hybrid fitness coaching method according to an embodiment of the present invention.

Referring to FIG. 5 , the personalized hybrid fitness coaching method according to an embodiment of the present invention includes collecting body information from the fitness center terminal 300 ( S410 ).

That is, a user who needs exercise can directly visit the fitness center offline to measure body information with a measuring device located in the fitness center, or a coach can directly check the user's physical condition.

When the fitness center terminal 300 is a measurement terminal such as a body composition measuring device located in a fitness center, sensing information of the fitness center terminal 300 is collected, and the fitness center terminal 300 is a coach (administrator) of the fitness center such as a computer. ), the information directly input by the coach may be collected.

When the user's body information is collected as described above, the server 100 of the present invention provides fitness coaching information that is optimal for the user's body information based on the database and algorithm built in advance as described in FIG. 2 (S420) .

In this case, the fitness coaching information provided to the user may be provided on the screen of the user terminal 200 such as a smart phone carried by the user through a program or an application.

The fitness coaching information may include customized fitness coaching information optimized for a user, nutritional diet information, and lifestyle information.

The user may exercise at home or another place based on the fitness coaching information provided to the user terminal 200 or change a nutritional diet and lifestyle.

The server 100 continuously collects body information from the user terminal 200 (S430). In this case, the body information collected from the user terminal 200 may be body-related information input by the user into the user terminal 200 , but may preferably be information sensed by a sensor mounted in the user terminal 200 .

For example, when the user terminal 200 is a smartphone, the server 100 may collect GPS sensor information, acceleration sensor information, and gyro sensor information.

The server 100 may generate the user's body information and lifestyle information based on this (S440).

Of course, the user terminal 200 is not limited to a smart phone and may be a smart watch or other health wearable device. Although GPS sensor information, acceleration sensor information, and gyro sensor information have been mentioned as the sensed information above, information from other sensors installed in smart watches or wearable devices, such as heart rate information, may also be possible.

When the server 100 collects the body-related information as described above from the user terminal 200 in real time, it updates the coaching information based on the pre-built big data DB and artificial intelligence learning algorithm (S450), and the user terminal 200 ) may provide updated fitness coaching information (S460).

6 is a flowchart illustrating an image analysis and future image information provision method of a personalized hybrid fitness coaching method according to an embodiment of the present invention.

The data collection unit 110 of the server 100 may collect the user's photographed image from the user terminal 200 or the fitness center terminal 300 (S510).

The image analysis unit 150 may analyze the collected image to analyze the user's body change information (S520).

The image generating unit 160 may generate future image information of the user based on the body change information ( S530 ), and the fitness recommendation unit 140 displays the user's future image information on the screen of a program or application provided to the user terminal 540 . of future image information may be provided (S540).

The personalized hybrid fitness coaching system and method of the present invention artificially learns the user's body information and lifestyle information with the above configuration and method, and provides online and offline fitness coaching using a model according to the learned result. Those of ordinary skill in the art to which the present invention pertains will understand that the present invention may be embodied in other specific forms without changing the technical spirit or essential features thereof. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

100: server
110: data collection unit
120: body state analysis unit
130: fitness coaching information generation unit
140: fitness coaching information provider
150: image analysis unit
160: image generator
170: user physical state DB
175: fitness model DB
177: fitness equipment DB
180: attribute matching DB
200: user terminal
300: fitness center terminal

Claims (11)

In the personalized hybrid fitness coaching system comprising a server connected to a network with a user terminal and a fitness center terminal,
The server is
a data collection unit for collecting body-related information of the user from the user terminal and the fitness center terminal;
a body condition analysis unit that analyzes the user's physical condition and lifestyle based on the body-related information;
Fitness coaching information for generating an inference model by learning the user's body-related information, the analyzed body state information, and lifestyle information with artificial intelligence, and generating fitness coaching information suitable for the user based on the generated inference model generator; and
A personalized hybrid fitness coaching system comprising a fitness coaching information provider for providing customized fitness coaching content and programs including the generated fitness coaching information, the user's current fitness status information and future fitness prediction information to the user terminal.
According to claim 1,
The fitness coaching information generator is a personalized hybrid fitness coaching system for generating the fitness coaching information by applying an artificial intelligence learning model to the body-related information that is updated based on a multi-armed bandit (MAB) algorithm.
3. The method of claim 2,
The fitness coaching information generator is a personalized hybrid fitness coaching system for generating a fitness goal achievement prediction probability and fitness prediction result based on the updated body-related information and pre-built big data information for each gender, age, and body type.
4. The method of claim 3,
The fitness coaching information generation unit generates and combines a plurality of machine learning or deep learning-based prediction models through ensemble learning, and using this, a personalized hybrid fitness coaching system for generating the fitness goal achievement prediction probability and fitness prediction result.
4. The method of claim 3,
The body-related information includes the user's photographed image information,
an image analysis unit extracting feature points from the photographed image information to analyze the user's body change information; and
Further comprising an image generator for generating image information of the user for the future within the fitness target period based on the analyzed body change information,
The fitness coaching information providing unit personalized hybrid fitness coaching system for providing the user's current image information and future image information generated by the image generation unit to the user terminal.
4. The method of claim 3,
The body-related information includes GPS sensor information, acceleration sensor information, and gyro sensor information received from the user terminal,
The body condition analyzer calculates the user's movement information based on the GPS sensor information, the acceleration sensor information, and the gyro sensor information to analyze the user's physical condition and lifestyle.
3. The method of claim 2,
The fitness coaching information is a personalized hybrid fitness coaching system including aerobic exercise information, anaerobic exercise information and FITT (frequency, intensity, time, form) information.
(a) collecting the user's body-related information from the user terminal and the fitness center terminal;
(b) analyzing the user's physical condition and lifestyle based on the body-related information;
(c) generating an inference model by learning the user's body-related information, the analyzed body state information, and lifestyle information with artificial intelligence, and generating fitness coaching information suitable for the user based on the generated inference model step; and
(d) personalized hybrid fitness coaching method comprising the step of providing customized coaching content and programs including the generated fitness coaching information, the user's current fitness status information and future fitness prediction information to a user terminal.
9. The method of claim 8,
The step (c) includes generating the fitness coaching information by applying an artificial intelligence learning model to the body-related information that is updated based on a multi-armed bandit (MAB) algorithm. coaching method.
10. The method of claim 9,
The step (c) includes generating a fitness goal achievement prediction probability and fitness prediction result based on the updated body-related information and pre-built big data information for each gender, age, and body type. Way.
11. The method of claim 10,
The body-related information includes the user's photographed image information,
The step (b) includes extracting feature points from the photographed image information and analyzing the user's body change information,
Step (c) is,
generating image information of a user for the future within a fitness target period based on the analyzed body change information; and
Personalized hybrid fitness coaching method comprising the step of providing the user's current image information and the generated future image information to the user terminal.

Images