TWI704943B - Artificial intelligence-assisted adaptive fitness training method - Google Patents

Abstract

An artificial intelligence-assisted adaptive fitness training method includes a pre-use real-time detection step, a weight calculation step, a pre-use determination step, a setting step, a post-use real-time detection step, an achievement-rate calculation step, and an adjustment determination step. The weight calculation step includes: using a processor to extract a user's normal physiological information from a cloud database; and after the processor receives a determined physiological information through a wearable device, using the processor to obtain a training weight according to a change value between the determined physiological information and the normal physiological information. The pre-use determination step is to use the processor to analyze whether the user can use a fitness device based on the training weight.

Description

A method of adaptive fitness training assisted by artificial intelligence

The present invention relates to a method for assisting adaptive fitness training with artificial intelligence, and particularly refers to a method for assisting adaptive fitness training with artificial intelligence that can set a fitness equipment so that users can use the fitness equipment appropriately according to real-time physical conditions.

With the advancement of medical technology, countries with high-tech development are gradually entering an aging society, and the number of elderly people will increase. Therefore, the health of the elderly is more important, and there is no doubt that a healthy life It must come from a balanced diet, proper and moderate exercise, and a happy mood. Among them, the use of fitness equipment indoors allows the elderly to maintain good exercise habits without worrying about the weather, and can also select suitable fitness equipment for parts of the body that need special strengthening.

However, there are many types of fitness equipment, and the operating conditions of each fitness equipment will affect the exercise effect of the user. Proper use can exert the best exercise effect. And the daily physical condition of the elderly will change. If the elderly do not Use fitness equipment according to the immediate physical condition, it is prone to sports injury or danger, and there may also be insufficient exercise.

In view of this, the inventor of the present invention proposes a method of using artificial intelligence to assist adaptive fitness training in order to enable any user to use a fitness equipment appropriately according to the immediate physical condition.

A method of using artificial intelligence to assist adaptive fitness training includes a step of using real-time detection, a step of calculating weights, a step of judging use, a setting step, a step of real-time detection after use, a step of calculating achievement rate, and a step of judging and adjusting .

The use of real-time detection step is to use a wearable device to detect a user's current real-time physiological information, so as to generate a judgment physiological information. The step of calculating the weight is to use a processor to retrieve a normal physiological information of the user from a cloud database, the processor then receives the determined physiological information through the wearable device, and compares the determined physiological information with respect to the The change value of the normal physiological information information is used to obtain a training weight. The determining use step is to use the processor to analyze whether the user can use a fitness equipment according to the training weight. The setting step is that when the processor analyzes that the user can use the fitness equipment, the processor retrieves a training condition of the user from the cloud database, and sets the fitness equipment according to the training weight and the training condition An operating condition of, where the training condition is an upper limit of the operating condition. The step of real-time detection after use is to use the wearable device to detect the real-time physiological information of the user after using the fitness equipment, so as to generate a physiological information of use. The step of calculating the achievement rate is to use the processor to receive from the fitness equipment an operation data of the user using the fitness equipment, and obtain an achievement rate according to the percentage of the operation data relative to the operating condition. The determining and adjusting step is to use the processor to analyze whether to adjust the training condition according to the achievement rate and the physiological information information received from the wearable device.

Further, the method of using artificial intelligence to assist adaptive fitness training further includes an identification step before the use of the real-time detection step. The identification step is to use an identifier to detect a physiological characteristic of the user and generate identification data.

Further, the method of assisting adaptive fitness training with artificial intelligence further includes a step of establishing a cloud database before the identification step, and the step of establishing a cloud database is to use the cloud database The identification data of the user, the normal physiological information information of the user in a normal state, and the training conditions related to the normal physiological information information are stored.

Further, the method of assisting adaptive fitness training with artificial intelligence further includes a cloud update step after the judgment adjustment step. The cloud update step is that when the processor adjusts the training condition, the processor adjusts the adjusted training condition Transfer to the cloud database, and use the cloud database to receive and update the adjusted training conditions.

Further, the method of assisting adaptive fitness training with artificial intelligence further includes a warning step after the step of calculating the attainment rate. The warning step is to compare the calculated attainment rate with a predetermined value by the processor, if the attainment rate is Below the predetermined value, the processor causes a reminder to generate a warning.

Further, the real-time physiological information includes at least one of a heartbeat value, a pulse value and a blood pressure value of the user.

Further, the operating condition includes at least one of a limited speed, a limited weight, a limited number of times, and a limited time.

According to the above technical features, the following effects can be achieved:

1. Through the use of real-time detection step and the calculation weight step, the processor obtains the training weight according to the judgment use step, analyzes whether the user can use the fitness equipment, and sets the fitness equipment in the setting step , When the user can use the fitness equipment, he can also do suitable fitness according to the real-time physiological condition. There is an upper limit setting, so that the user will not have sports injuries and accidents.

2. According to the step of calculating the achievement rate and the step of warning, if the achievement rate of the exercise of the user is lower than the predetermined value, in the case of insufficient exercise, the reminder generates the warning to remind the user.

3. By the step of real-time detection after use and the step of calculating the achievement rate, the processor obtains the physiological information of use and the achievement rate according to which the judgment adjustment step is performed, and the processor analyzes Whether to adjust the training conditions, so that the training conditions will be adjusted according to the user's condition, and always remain adaptable.

1: Cloud database

2: recognizer

3: wearable device

4: fitness equipment

41: Controller

5: Reminder

6: processor

S01: Steps to create a cloud database

S02: Identification step

S03: Use instant detection steps

S04: Steps to calculate weight

S05: Judging the use steps

S06: Setting procedure

S07: Steps for immediate detection after use

S08: Steps to calculate the achievement rate

S09: Warning steps

S10: Judging the adjustment steps

S11: Cloud update steps

[The first figure] is a flow chart showing a block diagram of an embodiment of the method for assisting adaptive fitness training with artificial intelligence in the present invention.

[Second Figure] is a flow chart illustrating a setting step and a real-time detection step after use in this embodiment.

[Figure 3] is a block diagram, a block diagram of a fitness system implementing this embodiment.

Based on the above technical features, the main effects of the method of the present invention using artificial intelligence to assist adaptive fitness training will be clearly presented in the following embodiments.

Please refer to the first to third figures, an embodiment of the method of the present invention for assisting adaptive fitness training with artificial intelligence is executed by a fitness system that includes a cloud database (1) and an identifier (2) ), a wearable device (3), a fitness equipment (4), a reminder (5) and a processor (6).

The cloud database (1) stores identification data of multiple users, and each identification data corresponds to a normal physiological information and a training condition. Among them, the identification data are respectively related to the corresponding physiological characteristics of the users, such as fingerprints, faces, irises, etc. of the users. Each piece of normal physiological information includes at least one of a body composition analysis, a muscle fat analysis, an obesity analysis, an abdominal obesity analysis, a heartbeat value, and a blood pressure value corresponding to the user in the normal state. Therefore, , Each training condition record is analyzed from the normal physiological information of each user to analyze the suitable sports equipment (4) and the limits of suitable use of the sports equipment (4), such as analyzing the normal physiological information of a user According to the information, the user's suitable limit for using a leg extension training machine is 15 times with a weight of 4.5 kg.

The recognizer (2) is arranged on the fitness equipment (4), and the recognizer (2) is used to perform recognition such as fingerprint recognition, face recognition, and iris recognition.

After being worn by a user, the wearable device (3) is used for detecting a real-time physiological information of the user. The real-time physiological information includes at least one of a heartbeat value, a pulse value and a blood pressure value of the user.

The fitness equipment (4) can be used by the users in turn. The fitness equipment (4) can be, for example, a treadmill, a bicycle, an elliptical machine, a flywheel, a rowing machine, a leg extension training machine, a leg extension machine, and the like.

The reminder (5) is used for generating a warning, and the warning can be a sound warning, a light warning or a picture warning.

The processor (6) is connected to the cloud database (1) via the Internet, and wirelessly connected to the wearable device (3), and electrically connected to the identifier (2) and a control inside the fitness equipment (4) (41), and the reminder (5).

The method of using artificial intelligence to assist adaptive fitness training includes a step of establishing a cloud database (S01), a step of identifying (S02), a step of using real-time detection (S03), a step of calculating weights (S04), and a step of determining use (S05), a setting step (S06), an immediate detection step after use (S07), a calculation achievement rate step (S08), a warning step (S09), a judgment adjustment step (S10), and a cloud update step (S11).

The step of creating a cloud database (S01) is to use the cloud database (1) to store the identification data of the users, the normal physiological information information of the users in the normal state, and the corresponding normal physiological information respectively Such training conditions for information. Therefore, the use of the fitness equipment (4) The user needs to first store the personal identification data, the normal physiological information, and the training conditions in the cloud database (1).

The identification step (S02) is to use the identifier (2) to detect the physiological characteristic of the user and generate the identification data.

The use real-time detection step (S03) is to use the wearable device (3) to detect the current real-time physiological information of the user, so as to generate a judgment physiological information information. That is, before the user starts to use the fitness equipment (4), the user wears the wearable device (3) to detect the current real-time physiological information.

The weight calculation step (S04) is to use the processor (6) to receive the identification data from the recognizer (2), and then retrieve the normal physiological information of the user from the cloud database (1) according to the identification data Information, the processor (6) receives the judgment physiological information information via the wearable device (3), and obtains a training weight based on the change value of the judgment physiological information information relative to the normal physiological information information.

The judgment use step (S05) is to use the processor (6) to analyze whether the user can use the fitness equipment (4) according to whether the training weight is greater than a minimum training weight, if the training weight is greater than the training weight The processor (6) analyzes that the user can use the fitness equipment (4), and the processor (6) then uses the setting step (S06). If the training weight is less than the minimum training weight, the processing The processor (6) analyzes that the user cannot use the fitness equipment (4), then the processor (6) does not execute the setting step (S06), and the user cannot use the fitness equipment (4). For example, the heartbeat value, the pulse value and the blood pressure value of the user's current real-time physiological information are the same as those in the normal state, the training weight calculated by the processor (6) is greater than the minimum training weight, and the processor (6) ) Analyze that the user can currently use the fitness equipment (4). If the user does not sleep or eat enough, resulting in a poor condition, so that the heartbeat value, the pulse value and the blood pressure value of the real-time physiological information are different from the normal state, the processor (6) judges the physiological information according to the Capital According to the change value of the normal physiological information information, the training weight is calculated to be less than the minimum value of the training weight, and the processor (6) analyzes that the user cannot currently use the fitness equipment (4).

The setting step (S06) is that when the processor (6) analyzes that the user can use the fitness equipment (4), the processor (6) then retrieves the user’s data from the cloud database (1) Training condition, and set an operating condition of the controller (41) of the fitness equipment (4) according to the training weight and the training condition, wherein the training condition is an upper limit of the operating condition, that is, if the When the user uses the fitness equipment (4) to exceed the limit of the operating conditions, the fitness equipment (4) will automatically stop, and the user is not allowed to continue using it. For example, according to the user’s normal physiological information, the normal limit of using the leg extension training machine is 15 times using a weight of 4.5 kg. This is the training condition. If the calculated training weight is 0.8, Then the operating conditions are 3.6 kg load 12 times. Therefore, after the processor (6) automatically sets the operating condition, if the user wants to make a limit that exceeds the operating condition, the fitness equipment (4) stops, and the user cannot use the fitness equipment (4) to do Training that exceeds the current physical load can prevent possible sports injuries or accidents.

The instant detection step (S07) after use is that after the user uses the fitness equipment (4), the user wears the wearable device (3), and uses the wearable device (3) to detect that the user uses the fitness equipment (4) The real-time physiological information afterwards is used to generate a physiological information information.

The step of calculating the achievement rate (S08) is to use the processor (6) to receive from the fitness equipment (4) an operation data of the user using the fitness equipment (4), and according to the operation data relative to the operating condition Percentage to get an achievement rate. For example, the processor (6) sets the operating condition of the leg extension training machine to load 3.6 kg for 12 times, and the user actually operates the leg extension training machine to load 3.6 kg for 8 times. Therefore, the operation The data is that the 3.6 kg load is 8 times, the achievement rate is 66.7%.

The warning step (S09) is to use the processor (6) to compare the calculated achievement rate with a predetermined value. If the achievement rate is lower than the predetermined value, the processor (6) causes the reminder (5) to generate One warning.

The judgment and adjustment step (S10) is to use the processor (6) to analyze whether to adjust the training condition according to the achievement rate and the physiological information information received from the wearable device (3). For example, the user’s fitness achievement rate has been low, and the physiological information used after exercise indicates that the user’s physiological state exceeds the normal load. The processor (6) analyzes that the training condition should be adjusted down. If the fitness achievement rate of the user has always reached 100%, and the physiological information used after exercise indicates that the user has not exceeded the normal load, the processor (6) analyzes that the training condition should be increased. If the processor (6) analyzes that the training condition does not need to be adjusted, the training condition is maintained. If the processor (6) analyzes that the training condition needs to be adjusted, skip to the cloud update step (S11).

The cloud update step (S11) is that when the processor (6) adjusts the training condition, the processor (6) transmits the adjusted training condition to the cloud database (1), and uses the cloud database (1) ) Receive the adjusted training condition and update it. When the user wants to use the fitness equipment (4) next time, the training condition retrieved by the processor (6) from the cloud database (1) is the update The training conditions that have been used. It should be added that if the training conditions are constantly updated, the processor (6) can make a display (not shown in the figure) remind the user to update the normal physiological information of the cloud database (1).

The whole process of the user using the fitness equipment (4) is: using the identifier (2) to detect the physiological characteristics of the user and generate the identification data, the user does not need to use the identifier (2) Enter the password, and then the user wears the wearable device (3) to detect the real-time physiological information, the processor (6) automatically calculates the training weight, and analyzes whether the user can use the fitness equipment (4), if the If the user can use the fitness equipment (4), the processor (6) will set the operating conditions of the fitness equipment (4). If the user cannot use the fitness equipment (4), the processor (6) will not Set the operating conditions, therefore, use The user can use the fitness equipment (4) without setting, which can save the time to understand and set the fitness equipment (4) before using the sports equipment (4). After use, the processor (6) calculates the At the same time, the user wears the wearable device (3) to detect the real-time physiological information. If the user’s fitness achievement rate is less than the predetermined value, the reminder (5) generates the warning to remind the user not to be lazy , And finally the processor (6) analyzes whether to adjust the training condition.

Therefore, the artificial intelligence-assisted adaptive fitness training method implemented by the fitness system is quite adaptable. Different operating conditions can be set for each user, with limited settings, so that these users will not In the event of a sports injury, if the user fails to reach the required amount of exercise, the system will also have the warning to remind the user, so this case is especially suitable for the use of silver-haired people, and the user will not set the fitness equipment (4) It doesn't matter, the processor (6) will automatically set the operating conditions. The best thing is that the system will keep updating the corresponding training conditions for each user's condition, so that the system will always be adaptable.

Based on the description of the above-mentioned embodiments, when one can fully understand the operation and use of the present invention and the effects of the present invention, the above-mentioned embodiments are only preferred embodiments of the present invention, and the implementation of the present invention cannot be limited by this. The scope, that is, simple equivalent changes and modifications made according to the scope of the patent application of the present invention and the description of the invention, are all within the scope of the present invention.

S01: Steps to create a cloud database

S02: Identification step

S03: Use instant detection steps

S04: Steps to calculate weight

S05: Judging the use steps

S06: Setting procedure

S07: Steps for immediate detection after use

S08: Steps to calculate the achievement rate

S09: Warning steps

S10: Judging the adjustment steps

S11: Cloud update steps

Claims (8)

An artificial intelligence-assisted adaptive fitness training method includes the following steps: a real-time detection step: a wearable device is used to detect a user's current real-time physiological information to generate a judgment physiological information information; a weighting step: A processor is used to retrieve a normal physiological information of the user from a cloud database, the processor then receives the judged physiological information through the wearable device, and compares the judged physiological information with the normal physiological information according to the wearable device To obtain a training weight; a judging step: using the processor to analyze whether the user can use a fitness equipment according to the training weight; a setting step: when the processor analyzes that the user can use the fitness When using the equipment, the processor retrieves a training condition of the user from the cloud database, and sets an operating condition of the fitness equipment according to the training weight and the training condition, wherein the training condition is one of the operating conditions Upper limit value; a step of real-time detection after use: the wearable device is used to detect the real-time physiological information of the user after using the fitness equipment, so as to generate a use of physiological information; a step of calculating the achievement rate: using the processor to obtain The fitness equipment receives an operation data of the user using the fitness equipment, and obtains an achievement rate according to the percentage of the operation data relative to the operating condition; and a judging adjustment step: using the processor according to the achievement rate and from the achievement rate The physiological information information received by the wearable device analyzes whether to adjust the training condition. As described in claim 1, the method for assisting adaptive fitness training with artificial intelligence further includes a recognition step before the use of the real-time detection step. The recognition step is to use a recognizer to detect a physiological characteristic of the user and generate 1. Identification data. As described in claim 2, the method for assisting adaptive fitness training with artificial intelligence further includes a step of creating a cloud database before the identification step, and the step of creating a cloud database is to use the cloud database to store the user’s The identification data, the normal physiological information information of the user in the normal state, and the training condition related to the normal physiological information information. As described in claim 1, the method for assisting adaptive fitness training with artificial intelligence further includes a cloud update step after the judgment adjustment step. The cloud update step is that when the processor adjusts the training condition, the processor will adjust The subsequent training conditions are transmitted to the cloud database, and the cloud database is used to receive and update the adjusted training conditions. As described in claim 1, the method for assisting adaptive fitness training with artificial intelligence further includes a warning step after the step of calculating the attainment rate, and the warning step is to compare the calculated attainment rate with a predetermined value by the processor If the achievement rate is lower than the predetermined value, the processor causes a reminder to generate a warning. The method for adaptive fitness training assisted by artificial intelligence as described in claim 1, wherein the real-time physiological information includes at least one of a heartbeat value, a pulse value and a blood pressure value of the user. The method for assisting adaptive fitness training with artificial intelligence as described in claim 1, wherein the normal physiological information includes a body composition analysis, a muscle fat analysis, an obesity analysis, and an abdominal obesity analysis related to the user , At least one of a heartbeat value and a blood pressure value. The method for assisting adaptive fitness training with artificial intelligence as described in claim 1, wherein the operating condition includes at least one of a speed limit, a weight limit, a limit number of times, and a limit time.

Images