Method and device for generating a posture training plan based on intelligent sensing airbag

Abstract

The application discloses a posture training plan generation method and device based on an intelligent sensing air bag. The method is applied to the intelligent sensing air bag, the intelligent sensing air bag comprises a sensor group and at least one air bag, and the method comprises the following steps: collecting training information corresponding to a user; determining target weights corresponding to the training information; generating a first posture training plan of the user according to the training information and the target weights corresponding to the training information, wherein the first posture training plan is used for determining control parameters of the intelligent sensing air bag, and the control parameters of the intelligent sensing air bag are used for controlling the intelligent sensing air bag to perform inflation / deflation operations corresponding to the control parameters when the user uses the intelligent sensing air bag to perform posture training, so as to guide the user to adjust the posture of the user according to the first posture training plan. It can be seen that the application can improve the generation accuracy of the posture training plan and is beneficial to improving the posture adjustment accuracy of the user.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a method and apparatus for generating posture training plans based on intelligent sensing airbags. Background Technology

[0002] As modern people face increasing pressure from studying and working, they often participate in planned physical exercise activities to release stress or relieve physical pain, such as posture training to address their own posture problems.

[0003] However, in real life, although professionals can guide the development of posture training plans, if users cannot accurately express their training progress, it may affect the accuracy of the plan's creation or adjustment. This can lead to users being unable to accurately adjust their postures when following the plan, resulting in suboptimal training outcomes. Therefore, proposing a technical solution to improve the accuracy of posture training plan development and thus enhance the accuracy of user posture adjustments is crucial. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and device for generating posture training plans based on intelligent sensing airbags, which can improve the accuracy of posture training plan generation and help improve the accuracy of user posture adjustment.

[0005] To address the aforementioned technical problems, the first aspect of this invention discloses a method for generating a posture training plan based on an intelligent sensing airbag. The method is applied to an intelligent sensing airbag, which includes a sensor array and at least one airbag. The method includes:

[0006] Collect training information corresponding to the user. The training information includes one or more combinations of first user body information, posture recognition information and stage completion information. The posture recognition information includes the current posture of the user that has been identified. The stage completion information is used to indicate the training completion degree of the user in the current training stage.

[0007] Determine the target weights corresponding to the training information, wherein the target weights corresponding to the training information include the sub-weights corresponding to each type of training sub-information in the training information, and the sum of the sub-weights corresponding to all the training sub-information is equal to one;

[0008] Based on the training information and the target weights corresponding to the training information, a first posture training plan for the user is generated. The first posture training plan includes one or more combinations of planned training posture, training duration, training frequency, and expected degree of body change. The first posture training plan is used to determine the control parameters of the intelligent sensing airbag. The control parameters of the intelligent sensing airbag are used to control the intelligent sensing airbag to perform the inflation / deflation operation corresponding to the control parameters when the user uses the intelligent sensing airbag for posture training, so as to guide the user to adjust the user's posture according to the first posture training plan.

[0009] As an optional implementation, in the first aspect of the present invention, determining the target weights corresponding to the training information includes:

[0010] Obtain the preset weights corresponding to each type of training sub-information in the training information;

[0011] Obtain the user's training difficulty level, which represents the range of training difficulty the user can tolerate;

[0012] Based on the training difficulty level, a standard range corresponding to each training sub-information in the training information is determined. The standard range corresponding to the training sub-information is used to represent the range of values ​​that can be used as the standard value corresponding to the training sub-information under the training difficulty level.

[0013] For each type of training sub-information in the training information, determine whether the training sub-information exceeds the standard range corresponding to the training sub-information, and obtain the range judgment result of the training sub-information;

[0014] Based on the range judgment results of all the training sub-information in the training information, all the training sub-information is divided into a first sub-information set and a second sub-information set. The first sub-information set includes all training sub-information that exceeds the corresponding standard range, and the second sub-information set includes all training sub-information that does not exceed the corresponding standard range.

[0015] Based on predetermined weight adjustment conditions, the preset weights corresponding to each training sub-information in the first sub-information set are adjusted to obtain a first weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the first sub-information set.

[0016] Based on the first weight adjustment result, the preset weights corresponding to each training sub-information in the second sub-information set are adjusted to obtain the second weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the second sub-information set.

[0017] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0018] Based on the first posture training plan and the training difficulty level, the standard force application method corresponding to each planned training posture in the first posture training plan is determined. The standard force application method is matched with the training difficulty level. The standard force application method includes the force application part, the force application degree corresponding to the force application part, and the force application duration corresponding to the force application part. The force application part is used to indicate the body part that the user needs to exert force voluntarily when completing the planned training posture.

[0019] The control parameters of the intelligent sensing airbag are determined based on the first posture training plan and the standard force application method corresponding to each of the planned training postures in the first posture training plan.

[0020] As an optional implementation, in the first aspect of the present invention, the stage completion information includes the target training completion degree of the current training stage, the actual training completion degree of the current training stage, and the completion difference degree of the current training stage.

[0021] The actual training completion rate of the current training phase is determined in the following way:

[0022] Based on the user's posture recognition information, the current posture difference is determined. The current posture difference is used to represent the degree of difference between the user's current posture and the standard training posture corresponding to the current posture. The standard training posture corresponding to the current posture is one of the standard training postures in the current training stage.

[0023] The actual training completion rate of the current training phase is calculated based on the posture difference degree corresponding to all the standard training postures in the current training phase.

[0024] The completion difference of the current training phase is determined in the following way:

[0025] The completion difference of the current training stage is calculated based on the target training completion rate and the actual training completion rate.

[0026] As an optional implementation, in the first aspect of the present invention, the posture recognition information is determined in the following manner:

[0027] The motion sensing data of the user collected in real time is subjected to adaptive weighted fusion processing to obtain multi-sensor fusion data; wherein, the motion sensing data is obtained based on the sensor group, which includes a distance sensor, an angle sensor and a pressure sensor. The distance sensor is used to collect the lifting height sensing data of the user's legs and / or hips, the angle sensor is used to collect the lifting angle sensing data of the user's legs and / or hips, and the pressure sensor is used to collect the pressure value inside the airbag;

[0028] The multi-sensor fused data is input into a pre-trained convolutional neural network model to obtain the user's first pose recognition information;

[0029] By analyzing the motion image data corresponding to the motion sensing data, the user's second posture recognition information is obtained. The motion image data is obtained based on the camera device corresponding to the intelligent sensing airbag.

[0030] The user's posture recognition information is determined based on the user's first posture recognition information and the user's second posture recognition information.

[0031] As an optional implementation, in the first aspect of the present invention, after generating the user's first pose training plan based on the training information and the target weights corresponding to the training information, the method further includes:

[0032] The user's second body information is obtained. The user's second body information is used to represent the user's body information after posture training using the intelligent sensing airbag according to the first posture training plan. The user's body information includes one or more combinations of user weight, user height, user body fat percentage, spinal status, pelvic status and leg status.

[0033] Based on the first user's physical information and the second user's physical information, determine the actual degree of physical change of the user;

[0034] Based on the predetermined user training objectives, it is determined whether the actual degree of physical change of the user has reached the expected degree of physical change corresponding to the user training objectives. The user training objectives include one or more combinations of daily exercise, fitness shaping and posture adjustment.

[0035] When it is determined that the user's actual physical changes have not reached the expected physical changes corresponding to the user's training objectives, the duration of the user's physical changes not meeting the target is counted. The duration of the physical changes not meeting the target is used to indicate the duration during which the user's actual physical changes have not reached the expected physical changes.

[0036] Determine whether the duration of the body changes that did not meet the standard exceeds the threshold for the duration of non-compliance corresponding to the first posture training plan;

[0037] When it is determined that the duration of the body change that does not meet the standard exceeds the threshold of the duration of the failure to meet the standard corresponding to the first posture training plan, the first posture training plan is adjusted according to the actual degree of body change and the duration of the body change that does not meet the standard to obtain a second posture training plan.

[0038] As an optional implementation, in the first aspect of the present invention, the method further includes:

[0039] When it is determined that the user's actual physical change reaches the expected physical change corresponding to the user's training objective, the user's physical change trend is determined based on the user's actual physical change and the user's historical physical change. The historical physical change is the physical change determined based on the user's physical information before / after using the intelligent sensing airbag for posture training according to the historical posture training plan.

[0040] Determine whether the fluctuation range of the stated body change trend exceeds the fluctuation range threshold;

[0041] When it is determined that the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold, the first posture training plan and the user's historical posture training plan are analyzed to obtain the cause of the trend fluctuation. The cause of the trend fluctuation is used to indicate the reason why the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold.

[0042] Based on the cause of the trend fluctuation, the first posture training plan is adjusted to obtain a third posture training plan, so that after the user performs posture training using the intelligent sensing airbag according to the third posture training plan, the fluctuation amplitude of the user's body change trend is lower than the fluctuation amplitude threshold.

[0043] A second aspect of the present invention discloses a posture training plan generation device based on an intelligent sensing airbag, the device being applied to an intelligent sensing airbag, the intelligent sensing airbag comprising a sensor group and at least one airbag, the device comprising:

[0044] The acquisition module is used to acquire training information corresponding to the user. The training information includes one or more combinations of first user body information, posture recognition information and stage completion information. The posture recognition information includes the identified current posture of the user, and the stage completion information is used to indicate the training completion degree of the user in the current training stage.

[0045] The first determining module is used to determine the target weight corresponding to the training information, wherein the target weight corresponding to the training information includes the sub-weight corresponding to each type of training sub-information in the training information, and the sum of the sub-weights corresponding to all the training sub-information is equal to one;

[0046] The generation module is used to generate a first posture training plan for the user based on the training information and the target weights corresponding to the training information. The first posture training plan includes one or more combinations of planned training posture, training duration, training frequency, and expected degree of body change. The first posture training plan is used to determine the control parameters of the intelligent sensing airbag. The control parameters of the intelligent sensing airbag are used to control the intelligent sensing airbag to perform the inflation / deflation operation corresponding to the control parameters when the user uses the intelligent sensing airbag for posture training, so as to guide the user to adjust the user's posture according to the first posture training plan.

[0047] As an optional implementation, in a second aspect of the present invention, the specific method by which the first determining module determines the target weights corresponding to the training information includes:

[0048] Obtain the preset weights corresponding to each type of training sub-information in the training information;

[0049] Obtain the user's training difficulty level, which represents the range of training difficulty the user can tolerate;

[0050] Based on the training difficulty level, a standard range corresponding to each training sub-information in the training information is determined. The standard range corresponding to the training sub-information is used to represent the range of values ​​that can be used as the standard value corresponding to the training sub-information under the training difficulty level.

[0051] For each type of training sub-information in the training information, determine whether the training sub-information exceeds the standard range corresponding to the training sub-information, and obtain the range judgment result of the training sub-information;

[0052] Based on the range judgment results of all the training sub-information in the training information, all the training sub-information is divided into a first sub-information set and a second sub-information set. The first sub-information set includes all training sub-information that exceeds the corresponding standard range, and the second sub-information set includes all training sub-information that does not exceed the corresponding standard range.

[0053] Based on predetermined weight adjustment conditions, the preset weights corresponding to each training sub-information in the first sub-information set are adjusted to obtain a first weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the first sub-information set.

[0054] Based on the first weight adjustment result, the preset weights corresponding to each training sub-information in the second sub-information set are adjusted to obtain the second weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the second sub-information set.

[0055] As an optional implementation, in a second aspect of the present invention, the first determining module is further configured to determine a standard force application method corresponding to each planned training posture in the first posture training plan based on the first posture training plan and the training difficulty level, wherein the standard force application method matches the training difficulty level, and the standard force application method includes the force application part, the force application degree corresponding to the force application part, and the force application duration corresponding to the force application part, wherein the force application part is used to represent the body part that the user needs to consciously apply force when completing the planned training posture;

[0056] The first determining module is further configured to determine the control parameters of the intelligent sensing airbag based on the first posture training plan and the standard force application method corresponding to each planned training posture in the first posture training plan.

[0057] As an optional implementation, in a second aspect of the present invention, the stage completion information includes the target training completion degree of the current training stage, the actual training completion degree of the current training stage, and the completion difference degree of the current training stage.

[0058] The actual training completion rate of the current training phase is determined in the following way:

[0059] Based on the user's posture recognition information, the current posture difference is determined. The current posture difference is used to represent the degree of difference between the user's current posture and the standard training posture corresponding to the current posture. The standard training posture corresponding to the current posture is one of the standard training postures in the current training stage.

[0060] The actual training completion rate of the current training phase is calculated based on the posture difference degree corresponding to all the standard training postures in the current training phase.

[0061] The completion difference of the current training phase is determined in the following way:

[0062] The completion difference of the current training stage is calculated based on the target training completion rate and the actual training completion rate.

[0063] As an optional implementation, in a second aspect of the invention, the pose recognition information is determined in the following manner:

[0064] The motion sensing data of the user collected in real time is subjected to adaptive weighted fusion processing to obtain multi-sensor fusion data; wherein, the motion sensing data is obtained based on the sensor group, which includes a distance sensor, an angle sensor and a pressure sensor. The distance sensor is used to collect the lifting height sensing data of the user's legs and / or hips, the angle sensor is used to collect the lifting angle sensing data of the user's legs and / or hips, and the pressure sensor is used to collect the pressure value inside the airbag;

[0065] The multi-sensor fused data is input into a pre-trained convolutional neural network model to obtain the user's first pose recognition information;

[0066] By analyzing the motion image data corresponding to the motion sensing data, the user's second posture recognition information is obtained. The motion image data is obtained based on the camera device corresponding to the intelligent sensing airbag.

[0067] The user's posture recognition information is determined based on the user's first posture recognition information and the user's second posture recognition information.

[0068] As an optional implementation, in a second aspect of the invention, the apparatus further includes:

[0069] The acquisition module is used to acquire the user's second user body information after the generation module generates the user's first posture training plan based on the training information and the target weights corresponding to the training information. The user's second user body information is used to represent the user's body information after posture training using the intelligent sensing airbag according to the first posture training plan. The user body information includes one or more combinations of user weight, user height, user body fat percentage, spinal status, pelvic status and leg status.

[0070] The second determining module is used to determine the actual degree of physical change of the user based on the first user's physical information and the second user's physical information;

[0071] The judgment module is used to determine whether the actual degree of physical change of the user has reached the expected degree of physical change corresponding to the user's training purpose, which includes one or more combinations of daily exercise, fitness shaping and posture adjustment.

[0072] The statistics module is used to count the duration during which the user's actual physical changes do not meet the expected physical changes corresponding to the user's training objective when the judgment module determines that the user's actual physical changes do not meet the expected physical changes. The duration during which the physical changes do not meet the objective is used to represent the duration during which the user's actual physical changes do not meet the expected physical changes.

[0073] The judgment module is also used to determine whether the duration of the body change not meeting the standard exceeds the threshold of the duration of the non-compliance corresponding to the first posture training plan;

[0074] An adjustment module is used to adjust the first posture training plan to obtain a second posture training plan when the judgment module determines that the duration of the body change that does not meet the standard exceeds the threshold of the duration of the non-compliance corresponding to the first posture training plan.

[0075] As an optional implementation, in a second aspect of the present invention, the second determining module is further configured to determine the user's body change trend based on the user's actual body change and the user's historical body change when the judging module determines that the user's actual body change reaches the expected body change corresponding to the user's training objective. The historical body change is the body change determined based on the user's body information before / after using the intelligent sensing airbag for posture training according to the historical posture training plan.

[0076] The judgment module is also used to determine whether the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold.

[0077] The device further includes:

[0078] The analysis module, when the judgment module determines that the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold, analyzes the first posture training plan and the user's historical posture training plan to obtain the cause of the trend fluctuation. The cause of the trend fluctuation is used to indicate the reason why the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold.

[0079] The adjustment module is further configured to adjust the first posture training plan according to the cause of the trend fluctuation to obtain a third posture training plan, so that after the user performs posture training using the intelligent sensing airbag according to the third posture training plan, the fluctuation amplitude of the user's body change trend is lower than the fluctuation amplitude threshold.

[0080] A third aspect of the present invention discloses another posture training plan generation device based on intelligent sensing airbags, the device comprising:

[0081] Memory containing executable program code;

[0082] A processor coupled to the memory;

[0083] The processor calls the executable program code stored in the memory to execute the posture training plan generation method based on intelligent sensing airbag disclosed in the first aspect of the present invention.

[0084] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute the posture training plan generation method based on intelligent sensing airbags disclosed in the first aspect of the present invention.

[0085] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0086] In this embodiment of the invention, training information corresponding to the user is collected. The training information includes one or more combinations of first user body information, posture recognition information, and stage completion information. The posture recognition information includes the current posture of the user, and the stage completion information is used to indicate the user's training completion degree in the current training stage. Target weights corresponding to the training information are determined. The target weights corresponding to the training information include sub-weights corresponding to each type of training sub-information in the training information, and the sum of the sub-weights corresponding to all training sub-information is equal to one. Based on the training information and the target weights corresponding to the training information, a first posture training plan for the user is generated. The first posture training plan includes one or more combinations of planned training posture, training duration, training frequency, and expected degree of body change. The first posture training plan is used to determine the control parameters of the intelligent sensing airbag. The control parameters of the intelligent sensing airbag are used to control the intelligent sensing airbag to perform inflation / deflation operations corresponding to the control parameters when the user uses the intelligent sensing airbag for posture training, so as to guide the user to adjust the user's posture according to the first posture training plan. As can be seen, implementing this invention can collect the user's corresponding training information and determine the target weights corresponding to the training information, and generate the user's posture training plan based on the training information and its corresponding target weights. This can improve the accuracy of posture training plan generation, thereby improving the accuracy of posture training plan and the adaptability of posture training plan to user training conditions, thus improving the accuracy of user posture adjustment. Furthermore, it can determine the control parameters of the intelligent sensing airbag based on the posture training plan, which can improve the control accuracy of the intelligent sensing airbag, thereby further improving the accuracy of user posture adjustment and improving the user's experience with the intelligent sensing airbag. Attached Figure Description

[0087] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0088] Figure 1 This is a flowchart illustrating a method for generating a posture training plan based on an intelligent sensing airbag, as disclosed in an embodiment of the present invention.

[0089] Figure 2 This is a flowchart illustrating another method for generating a posture training plan based on an intelligent sensing airbag, as disclosed in an embodiment of the present invention.

[0090] Figure 3 This is a schematic diagram of the airbag structure of an intelligent sensing airbag, which is a posture training plan generation method based on an intelligent sensing airbag disclosed in an embodiment of the present invention.

[0091] Figure 4 This is a schematic diagram of the structure of an attitude training plan generation device based on an intelligent sensing airbag disclosed in an embodiment of the present invention;

[0092] Figure 5 This is a schematic diagram of another posture training plan generation device based on intelligent sensing airbags disclosed in an embodiment of the present invention.

[0093] Figure 6 This is a schematic diagram of another posture training plan generation device based on intelligent sensing airbag disclosed in an embodiment of the present invention. Detailed Implementation

[0094] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0095] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0096] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0097] This invention discloses a method and apparatus for generating posture training plans based on intelligent sensing airbags. It can collect user-specific training information and determine the target weights corresponding to that information. Based on the training information and its corresponding target weights, it generates a posture training plan for the user, improving the accuracy of posture training plan generation and thus enhancing the fit between the posture training plan and the user's training situation, thereby improving the accuracy of user posture adjustment. Furthermore, it can determine the control parameters of the intelligent sensing airbag based on the posture training plan, improving the control accuracy of the intelligent sensing airbag and further enhancing the accuracy of user posture adjustment, ultimately improving the user experience with the intelligent sensing airbag. These are described in detail below.

[0098] Example 1

[0099] Please see Figure 1 , Figure 1 This is a flowchart illustrating a posture training plan generation method based on intelligent sensing airbags disclosed in an embodiment of the present invention. Figure 1 The described posture training plan generation method based on intelligent sensing airbags can be applied to intelligent sensing airbags, which include a sensor group and at least one airbag. When the number of airbags is eight, the schematic diagram of the airbag structure of the intelligent sensing airbag can be shown as follows. Figure 3 As shown, this method can also be applied to one of the following: a device, terminal, system, and server for generating posture training plans corresponding to intelligent sensing airbags. The server includes a cloud server or a local server; this embodiment of the invention does not limit the specific application. Figure 1 As shown, the posture training plan generation method based on intelligent sensing airbags may include the following operations:

[0100] 101. Collect the training information corresponding to the user.

[0101] In this embodiment of the invention, the training information may include one or more combinations of first user body information, posture recognition information, and stage completion information. The user is the subject of posture training using intelligent sensing airbags; the user body information may include one or more combinations of user weight, user body fat percentage, spinal status, pelvic status, and leg status; the posture recognition information may include the identified current posture of the user; the stage completion information indicates the user's training completion rate in the current training stage, and the current training stage may include a duration of posture training currently being performed by the user and at least one standard training posture that the user needs to complete within that duration.

[0102] 102. Determine the target weights corresponding to the training information.

[0103] In this embodiment of the invention, the target weight corresponding to the training information may include the sub-weight corresponding to each type of training sub-information in the training information, and the sum of the sub-weights corresponding to all training sub-information is equal to one.

[0104] 103. Generate the user's first posture training plan based on the training information and the target weights corresponding to the training information.

[0105] In this embodiment of the invention, the first posture training plan may include one or more combinations of planned training postures, training duration, training frequency, and expected degree of body change. The first posture training plan is used to determine the control parameters of the intelligent sensing airbag. These control parameters are used to control the intelligent sensing airbag to perform inflation / deflation operations corresponding to the control parameters when the user uses it for posture training, thereby guiding the user to adjust their posture according to the first posture training plan. The training duration may include the duration of a single training session and / or the duration of a training cycle; the training frequency may include the frequency of completing a single planned training posture and / or the frequency of completing a combination of planned training postures; and the expected degree of body change is used to represent the degree of change in the user's body information after posture training according to the first posture training plan. For example, the duration of a single training session in the posture training plan is 10 minutes, and the training cycle lasts for one week. The posture training plan includes posture A, posture B, and posture C. In a single posture training session, the user needs to complete 5 repetitions of posture A, 4 repetitions of posture B, and 2 repetitions of posture C. The user needs to complete 3 posture training sessions per day. After the training cycle, the user's expected physical change is a weight loss of 1 kg compared to before the posture training.

[0106] As can be seen, the method described in the embodiments of the present invention can collect the user's corresponding training information and determine the target weights corresponding to the training information, and generate the user's posture training plan based on the training information and its corresponding target weights. This can improve the accuracy of posture training plan generation, thereby improving the accuracy of posture training plan and improving the adaptability of posture training plan to user training conditions, thus improving the accuracy of user posture adjustment. Furthermore, it can determine the control parameters of the intelligent sensing airbag based on the posture training plan, which can improve the control accuracy of the intelligent sensing airbag, thereby further improving the accuracy of user posture adjustment and improving the user's experience with the intelligent sensing airbag.

[0107] In an optional embodiment, the phase completion information may include the target training completion rate of the current training phase, the actual training completion rate of the current training phase, and the completion difference rate of the current training phase.

[0108] The actual training completion rate in the current training phase can be determined in the following ways:

[0109] Based on the user's corresponding posture recognition information, the current posture difference is determined. The current posture difference is used to represent the degree of difference between the user's current posture and the standard training posture corresponding to the current posture. The standard training posture corresponding to the current posture is one of the standard training postures in the current training stage.

[0110] The actual training completion rate of the current training phase is calculated based on the posture difference rate corresponding to all standard training postures in the current training phase.

[0111] The completion difference in the current training phase can be determined in the following ways:

[0112] The completion difference of the current training stage is calculated based on the target training completion rate and the actual training completion rate.

[0113] The completion difference of the current training phase is used to represent the degree of difference between the actual training completion and the target training completion.

[0114] Optionally, the actual training completion rate of the current training phase can be calculated by determining the completion rate of each standard training posture based on the posture difference rate corresponding to all standard training postures in the current training phase, and then performing a weighted average of the completion rates of each standard training posture to obtain the actual training completion rate of the current training phase. Optionally, the completion difference rate of the current training phase can be the difference between the target training completion rate and the actual training completion rate.

[0115] For example, the degree of completion and the degree of completion difference in the embodiments of the present invention can be expressed as percentages.

[0116] As can be seen, this optional embodiment can determine the current posture difference degree based on the user's current posture and the standard training posture corresponding to the current posture, and calculate the actual training completion degree based on the posture difference degree corresponding to all standard training postures in the current training stage. This can improve the accuracy of determining the actual training completion degree, thereby improving the accuracy of stage completion information and thus improving the accuracy of posture training plan generation. Furthermore, it can calculate the completion difference degree based on the target training completion degree and the actual training completion degree, which can improve the accuracy of determining the completion difference degree and thus improve the accuracy of stage completion information.

[0117] In another alternative embodiment, the pose recognition information can be determined in the following way:

[0118] Adaptive weighted fusion processing is performed on the real-time user motion sensing data to obtain multi-sensor fusion data. The motion sensing data is obtained based on a sensor group, which includes a range sensor, an angle sensor, and a pressure sensor. The range sensor is used to collect the user's leg and / or hip lift height sensing data, the angle sensor is used to collect the user's leg and / or hip lift angle sensing data, and the pressure sensor is used to collect the pressure value inside the airbag.

[0119] The multi-sensor fusion data is input into a pre-trained convolutional neural network model to obtain the user's first pose recognition information;

[0120] By analyzing the motion image data corresponding to the motion sensing data, the user's second posture recognition information is obtained. The motion image data is obtained based on the camera device corresponding to the intelligent sensing airbag.

[0121] The user's posture recognition information is determined based on the user's first posture recognition information and the user's second posture recognition information.

[0122] Optionally, adaptive weighted fusion processing can be performed on the real-time collected motion sensing data of the user to obtain multi-sensor fused data. Specifically, this can be done by first filtering the motion sensing data using a filtering function (e.g., a third-order Butterworth function) to obtain posture information such as leg lift height, leg lift angle, hip lift height, hip lift angle, and airbag pressure value. Then, the posture information is normalized, and the temporal features of each normalized posture information are extracted. These temporal features can include the mean, standard deviation, and covariance corresponding to each normalized posture information. Finally, based on pre-defined multi-sensor data... The adaptive weighted fusion model performs adaptive weighted fusion processing on all time-domain features to obtain multi-sensor fused data. Optionally, when the adaptive weighted fusion model performs adaptive weighted fusion processing on all time-domain features, the characteristic linear function between each time-domain feature is first determined. Based on the characteristic linear function and the adaptive weighted fusion algorithm, the time-domain features are weighted and fused. The optimal weighting factor for each time-domain feature vector is obtained adaptively based on the condition of minimizing the total mean square error of the time-domain features. The time-domain features are then processed according to the optimal weighting factor to obtain multi-sensor fused data.

[0123] Optionally, motion image data corresponding to motion sensing data can be analyzed using image recognition algorithms to obtain the user's second posture recognition information.

[0124] In this embodiment, the motion image data and the corresponding motion sensing data are acquired at the same time. The motion image data may include motion image data from at least one shooting angle. The content of the motion image data may include the user's movement state and the user's facial expressions during movement. This embodiment of the invention does not impose any limitations.

[0125] As can be seen, this optional embodiment can perform fusion processing on the collected motion sensing data to obtain multi-sensor fusion data, and input the multi-sensor fusion data into a convolutional neural network model to obtain first posture recognition information, and analyze motion image data to obtain second posture recognition information. Based on the first posture recognition information and the second posture recognition information, the user's posture recognition information can be determined, which can improve the accuracy of posture recognition information determination, thereby improving the accuracy of posture recognition information and thus improving the accuracy of posture training plan generation.

[0126] In this embodiment of the invention, the intelligent sensing airbag may include eight airbags, an inflation / deflation unit, and a data acquisition unit. The inflation / deflation unit includes an air pump, a solenoid valve, and a corresponding switching device for the solenoid valve. The data acquisition unit includes a sensor group and an MCU. For example, a schematic diagram of the airbag structure of the intelligent sensing airbag based on the posture training plan generation method of the intelligent sensing airbag can be shown as follows: Figure 3As shown, the intelligent sensing airbag system comprises eight airbags, each with an independent chamber that is not interconnected; the airbags can be made of PVC (polyvinyl chloride); the eight airbags correspond to... Figure 3 The eight airbags shown (A1, A2, B1, B2, C1, C2, D1, and D2) have the following dimensions and arrangement: Figure 3 As shown, the volumes of airbags A1, A2, C1, and C2 are 1.5L, and the volumes of airbags B1 and B2 are 0.5L. Figure 3 The air outlets ①, ②, ③, and ④ in the middle correspond to the solenoid valves of the four airbags on the left, and the air outlets ⑤, ⑥, ⑦, and ⑧ correspond to the solenoid valves of the four airbags on the right. Figure 3 I, II, and III are three different types of air tubes. I is an air tube with a pressure gauge, II can inflate a single airbag, and III can inflate two airbags simultaneously. By changing the inflation volume of the airbags, the user's lower back and hips can be raised to control the height of their lower back and hips off the bed. When the user completes the "bridge" and "bow" poses, the solenoid valves inflate a total of six airbags (A1, A2, B1, B2, C1, and C2), allowing the user to raise their lower back and hips with the assistance of the airbags. D1 and D2 are auxiliary airbags; inflating D1 and D2 simultaneously can raise the user's lower limbs.

[0127] Example 2

[0128] Please see Figure 2 , Figure 2 This is a flowchart illustrating a posture training plan generation method based on intelligent sensing airbags disclosed in an embodiment of the present invention. Figure 2 The described posture training plan generation method based on intelligent sensing airbags can be applied to intelligent sensing airbags, which include a sensor group and at least one airbag. When the number of airbags is eight, the schematic diagram of the airbag structure of the intelligent sensing airbag can be shown as follows. Figure 3 As shown, this method can also be applied to one of the following: a device, terminal, system, and server for generating posture training plans corresponding to intelligent sensing airbags. The server includes a cloud server or a local server; this embodiment of the invention does not limit the specific application. Figure 2 As shown, the posture training plan generation method based on intelligent sensing airbags may include the following operations:

[0129] 201. Collect the training information corresponding to the user.

[0130] 202. Obtain the preset weights corresponding to each type of training sub-information in the training information.

[0131] In this embodiment of the invention, the preset weights corresponding to the training sub-information can be obtained by averaging the preset weights corresponding to all training sub-information based on the number of training sub-information, or they can be weights predetermined based on the information type of the training sub-information. This embodiment of the invention does not limit the weights.

[0132] 203. Obtain the user's training difficulty level.

[0133] In this embodiment of the invention, the training difficulty level is used to represent the range of training difficulty that a user can tolerate. Optionally, the specific method for obtaining a user's training difficulty level can be: determining the user's training difficulty level based on the user's basic information and user training records. The user's basic information may include one or more combinations of user gender, user age, and user height. The user training records may include one or more combinations of historical total training time, historical training difficulty levels, historical training postures, and historical training plans. This embodiment of the invention does not limit this. For example, assuming the user is female, 40 years old, and 1.65 meters tall, and the user has not undergone posture training, then the user's training difficulty level is determined to be easy.

[0134] 204. Based on the training difficulty level, determine the standard range corresponding to each training sub-information in the training information.

[0135] In this embodiment of the invention, the standard range corresponding to the training sub-information is used to represent the range of values ​​that can be used as the standard value corresponding to the training sub-information under the training difficulty level; for example, assuming the training difficulty level is easy, the standard range corresponding to the user's body fat percentage in the first user body information can be 28% to 30%, the current posture difference can be 60% to 70%, and the standard range corresponding to the completion difference of the current training stage in the stage completion information can be 60% to 65%.

[0136] 205. For each type of training sub-information in the training information, determine whether the training sub-information exceeds the standard range corresponding to the training sub-information, and obtain the range judgment result of the training sub-information.

[0137] In this embodiment of the invention, the range judgment result of the training sub-information can be used to indicate whether the training sub-information exceeds the standard range corresponding to the training sub-information or does not exceed the standard range corresponding to the training sub-information.

[0138] 206. Based on the range judgment results of all training sub-information in the training information, divide all training sub-information into the first sub-information set and the second sub-information set.

[0139] In this embodiment of the invention, the first sub-information set may include all training sub-information that exceeds the corresponding standard range, and the second sub-information set may include all training sub-information that does not exceed the corresponding standard range.

[0140] 207. Based on the predetermined weight adjustment conditions, adjust the preset weights corresponding to each training sub-information in the first sub-information set to obtain the first weight adjustment result.

[0141] In this embodiment of the invention, the first weight adjustment result may include the sub-weight corresponding to each training sub-information in the first sub-information set, wherein when neither the first sub-information set nor the second sub-information set is empty, the sub-weight corresponding to each training sub-information in the first sub-information set is higher than the preset weight corresponding to that training sub-information.

[0142] It should be noted that when the second sub-information set is empty, step 209 can be executed directly after step 207; and optionally, when the second sub-information set is empty, the difference between each training sub-information in the first sub-information set and the standard range corresponding to that training sub-information can be calculated based on the predetermined weight adjustment conditions to obtain the difference degree corresponding to each training sub-information, and the preset weight corresponding to each training sub-information in the first sub-information set can be adjusted according to the difference degree corresponding to each training sub-information to obtain the first weight adjustment result.

[0143] 208. Based on the first weight adjustment result, adjust the preset weights corresponding to each training sub-information in the second sub-information set to obtain the second weight adjustment result.

[0144] In this embodiment of the invention, the second weight adjustment result may include the sub-weights corresponding to each training sub-information in the second sub-information set.

[0145] For example, suppose the training information includes three training sub-information items ①, ②, and ③, where the preset weights corresponding to training sub-information items ①, ②, and ③ are 0.2, 0.4, and 0.4, respectively. If it is determined that training sub-information item ① exceeds the corresponding standard range, while training sub-information items ② and ③ do not exceed the corresponding standard range, based on the preset weight adjustment conditions, the preset weight corresponding to training sub-information item ① can be proportionally increased to 0.4, while the preset weights corresponding to training sub-information items ② and ③ are decreased to 0.3 and 0.3, respectively, along with the weight corresponding to training sub-information item ①.

[0146] 209. Generate the user's first pose training plan based on the training information and the target weights corresponding to the training information.

[0147] For further detailed descriptions of steps 201 and 209 in this embodiment of the invention, please refer to the detailed descriptions of steps 101 and 103 in Embodiment 1. These descriptions will not be repeated in this embodiment of the invention.

[0148] As can be seen, the method described in the embodiments of the present invention can collect the user's corresponding training information and determine the target weights corresponding to the training information, and generate the user's posture training plan based on the training information and its corresponding target weights. This can improve the accuracy of posture training plan generation, thereby improving the accuracy of posture training plan, which is conducive to improving the adaptability of posture training plan to user training situation, and thus improving the accuracy of user posture adjustment. Furthermore, it can determine the control parameters of intelligent sensing airbag based on posture training plan, which can improve the control accuracy of intelligent sensing airbag, thereby further improving the accuracy of user posture adjustment, and thus improving the user's experience with intelligent sensing airbag. Furthermore, based on the acquired training difficulty level, it can determine the standard range corresponding to each training sub-information in the training information, and determine whether the training sub-information exceeds the corresponding standard range to obtain the range judgment result. Based on the range judgment result, it divides all training sub-information into a first sub-information set and a second sub-information set. After adjusting the preset weights corresponding to each training sub-information in the first sub-information set to obtain the first weight adjustment result, it adjusts the second sub-information set based on the first weight adjustment result to obtain the second weight adjustment result. This can improve the accuracy of determining the standard range corresponding to the training sub-information, thereby improving the accuracy of dividing the sub-information set, and further improving the accuracy of determining the sub-weights corresponding to the training sub-information, which is conducive to improving the accuracy of attitude training plan generation.

[0149] In an optional embodiment, the method may further include the following operations:

[0150] Based on the first posture training plan and the training difficulty level, the standard force application method corresponding to each planned training posture in the first posture training plan is determined. The standard force application method is matched with the training difficulty level. The standard force application method includes the force application part, the force application degree corresponding to the force application part, and the force application duration corresponding to the force application part. The force application part is used to indicate the body part that the user needs to exert force voluntarily when completing the planned training posture.

[0151] Based on the first posture training plan and the standard force application method corresponding to each planned training posture in the first posture training plan, the control parameters of the intelligent sensing airbag are determined.

[0152] The control parameters of the intelligent sensing airbag can include one or more combinations of inflation / deflation mode, inflation / deflation volume, inflation / deflation speed, and inflation / deflation frequency. Specifically, the intelligent sensing airbag can be controlled to perform inflation / deflation operations according to the control parameters of the intelligent sensing airbag to adjust the shape of the airbag. The shape of the airbag includes one or more combinations of the length, width, height, and intracavity air pressure value of the airbag. The user's training posture can be adjusted by adjusting the shape of the airbag.

[0153] For example, if the training difficulty level is easy, the user's effort can be reduced, and the air pressure inside the airbag can be increased to help the user complete the planned training posture. If the training difficulty level is hard, the user's effort can be increased, and the air pressure inside the airbag can be reduced to guide the user to complete the planned training posture independently.

[0154] As can be seen, this optional embodiment can determine the standard force application method corresponding to each planned training posture based on the generated posture training plan and training difficulty level, and determine the control parameters of the intelligent sensing airbag based on the posture training plan and the corresponding standard force application method. This can improve the accuracy of determining the standard force application method corresponding to the training posture, thereby improving the accuracy of determining the control parameters of the intelligent sensing airbag, and further improving the control accuracy of the intelligent sensing airbag, which is conducive to improving the user's experience with the intelligent sensing airbag.

[0155] In another optional embodiment, after generating the user's first pose training plan based on the training information and the target weights corresponding to the training information, the method may further include the following operations:

[0156] The user's second body information is obtained. The user's second body information is used to represent the user's body information after posture training using intelligent sensing airbags according to the first posture training plan. The user's body information includes one or more combinations of user weight, user body fat percentage, spinal status, pelvic status and leg status.

[0157] Based on the physical information of the first user and the physical information of the second user, determine the actual degree of physical change of the user;

[0158] Based on the predetermined user training objectives, determine whether the user's actual physical changes have reached the expected physical changes corresponding to the user training objectives. User training objectives include one or more combinations of daily exercise, fitness shaping, and posture adjustment.

[0159] When it is determined that the user's actual physical changes have not reached the expected physical changes corresponding to the user's training objectives, the duration of the user's physical changes that have not met the standards is counted. The duration of the physical changes that have not met the standards is used to indicate the duration for which the user's actual physical changes have not reached the expected physical changes.

[0160] Determine whether the duration of unmet physical changes exceeds the threshold for unmet duration corresponding to the first posture training plan;

[0161] When it is determined that the duration of unmet physical change exceeds the threshold for unmet physical change in the first posture training plan, the first posture training plan is adjusted according to the actual degree of physical change and the duration of unmet physical change to obtain the second posture training plan.

[0162] Optionally, if it is determined that the duration of the body change failure does not exceed the failure duration threshold corresponding to the first posture training plan, the first posture training plan can remain unchanged.

[0163] The expected degree of physical change varies depending on the user's training objective. For example, the expected degree of physical change for a user whose training objective is daily exercise is lower than the expected degree of physical change for a user whose training objective is fitness and body shaping. The threshold for the duration of non-compliance corresponding to the first posture training plan can be determined based on the specific content included in the first posture training plan. Furthermore, the threshold for the duration of non-compliance corresponding to the first posture training plan can be determined based on the first posture training plan and the training difficulty level. For example, the higher the training difficulty level, the higher the threshold for the duration of non-compliance. This embodiment of the invention does not impose any limitations on this.

[0164] As can be seen, this optional embodiment can also determine the actual degree of physical change of the user after training according to the posture training plan. When it is determined that the actual degree of physical change of the user has not reached the expected degree of physical change corresponding to the user's training purpose, the duration of the physical change not meeting the standard is counted. If the duration of the physical change not meeting the standard exceeds the threshold of the duration of the non-compliance corresponding to the posture training plan, the posture training plan is adjusted according to the actual degree of physical change and the duration of the non-compliance. This can improve the accuracy of determining the degree of physical change of the user, thereby improving the accuracy of judging whether the degree of physical change of the user has met the standard, and further improving the accuracy of adjusting the posture training plan. This is beneficial to reducing the possibility of user injury during training and improving the accuracy of user posture adjustment.

[0165] In this optional embodiment, the method may also include the following operations:

[0166] When it is determined that the user's actual physical change has reached the expected physical change corresponding to the user's training objective, the user's physical change trend is determined based on the user's actual physical change and the user's historical physical change. The historical physical change is the physical change determined based on the user's physical information before / after using intelligent sensing airbags for posture training according to the historical posture training plan.

[0167] Determine whether the fluctuation range of the body's change trend exceeds the fluctuation range threshold;

[0168] When it is determined that the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold, the first posture training plan and the user's historical posture training plan are analyzed to obtain the cause of the trend fluctuation. The cause of the trend fluctuation is used to indicate the reason why the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold.

[0169] Based on the reasons for the trend fluctuations, the first posture training plan is adjusted to obtain the third posture training plan, so that after the user uses the intelligent sensing airbag for posture training according to the third posture training plan, the fluctuation amplitude of the user's body change trend is lower than the fluctuation amplitude threshold.

[0170] The user's body change trend can be used to represent the change in the user's body information from the initial moment of posture training to the current moment, and the user's body change trend can be displayed through line graphs, histograms, etc. If the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold, it indicates that the user's physical condition is unstable. Furthermore, it can indicate that the results of the user's posture adjustment are repeatedly changing, and the posture training plan is not suitable for the user, requiring adjustment of the posture training plan. Specifically, the training difficulty level, the training frequency and / or training posture in the posture training plan can be adjusted. Furthermore, the control parameters of the intelligent sensing airbag can be adjusted according to the adjusted training difficulty level. This embodiment of the invention is not limited to this.

[0171] As can be seen, this optional embodiment can also determine whether the fluctuation range of the body change trend exceeds the fluctuation range threshold when it is determined that the user's actual body change has reached the expected body change corresponding to the user's training purpose. If it exceeds the threshold, the posture training plan and historical posture training plans are analyzed to find the reason for the excessive fluctuation range of the body change trend. The posture training plan is then adjusted according to the reason for the trend fluctuation. This can improve the accuracy of determining the user's body change trend, thereby improving the flexibility and accuracy of adjusting the posture training plan. It is beneficial to improve the adaptability between the posture training plan and the user's body change trend, and thus improve the accuracy of adjusting the user's posture.

[0172] In this optional embodiment, the method may further include the following operations:

[0173] When it is determined that the fluctuation range of the body change trend does not exceed the fluctuation range threshold, and when the body change trend is detected to indicate an upward trend, the user is instructed to continue to use the intelligent sensing airbag to perform posture training according to the first posture training plan.

[0174] When a trend of body change is detected to represent a downward trend, a target body sub-information in the second user's body information is determined, wherein the target body sub-information is used to represent body sub-information in the second user's body information that is below the standard based on the first user's body information;

[0175] Analyze the first posture training plan to obtain the reasons for the failure to meet the target body sub-information;

[0176] Based on the reasons for the failure of the target body sub-information to meet the standards, the first posture training plan is adjusted to obtain the fourth posture training plan.

[0177] For example, the reasons for the failure of the target body sub-information may be planning reasons such as too low training frequency and too short training duration, or airbag usage reasons such as incorrect use of the intelligent sensing airbag by the user (e.g., incorrect placement of the intelligent sensing airbag). This embodiment of the invention does not limit the scope of the invention.

[0178] As can be seen, this optional embodiment can also identify target body sub-information that is below standard when a downward trend in the user's body change is detected, and find out the reasons for the sub-information not meeting the standard by analyzing the posture training plan. Based on the reasons for the sub-information not meeting the standard, the posture training plan can be adjusted, which can improve the accuracy of identifying the factors in the posture training plan that cause the user's body information to not meet the standard, thereby improving the accuracy of the posture training plan adjustment and further improving the precision of the user's posture adjustment.

[0179] Example 3

[0180] Please see Figure 4 , Figure 4 This is a schematic diagram of a posture training plan generation device based on an intelligent sensing airbag, as disclosed in an embodiment of the present invention. Figure 4 The described attitude training plan generation device based on intelligent sensing airbags can be applied to intelligent sensing airbags, which include a sensor group and at least one airbag. When the number of airbags is eight, the schematic diagram of the airbag structure of the intelligent sensing airbag is as follows: Figure 3 As shown, this device can also be applied to one of the posture training plan generation devices, terminals, systems, and servers corresponding to intelligent sensing airbags, wherein the server includes a cloud server or a local server, and the embodiments of the present invention are not limited thereto. Figure 4 As shown, the posture training plan generation device based on intelligent sensing airbags may include:

[0181] The acquisition module 301 is used to acquire training information corresponding to the user. The training information includes one or more combinations of the first user's body information, posture recognition information and stage completion information. The posture recognition information includes the current posture of the user that has been identified. The stage completion information is used to indicate the user's training completion degree in the current training stage.

[0182] The first determining module 302 is used to determine the target weights corresponding to the training information. The target weights corresponding to the training information include the sub-weights corresponding to each type of training sub-information in the training information, and the sum of the sub-weights corresponding to all training sub-information is equal to one.

[0183] The generation module 303 is used to generate a first posture training plan for the user based on the training information and the target weights corresponding to the training information. The first posture training plan includes one or more combinations of planned training posture, training duration, training frequency and expected degree of body change. The first posture training plan is used to determine the control parameters of the intelligent sensing airbag. The control parameters of the intelligent sensing airbag are used to control the intelligent sensing airbag to perform the inflation / deflation operation corresponding to the control parameters when the user uses the intelligent sensing airbag for posture training, so as to guide the user to adjust the user's posture according to the first posture training plan.

[0184] As can be seen, the device described in the embodiments of the present invention can collect the user's corresponding training information and determine the target weights corresponding to the training information, and generate the user's posture training plan based on the training information and its corresponding target weights. This can improve the accuracy of posture training plan generation, thereby improving the accuracy of posture training plan and improving the adaptability of posture training plan to user training conditions, thus improving the accuracy of user posture adjustment. Furthermore, it can determine the control parameters of the intelligent sensing airbag based on the posture training plan, which can improve the control accuracy of the intelligent sensing airbag, thereby further improving the accuracy of user posture adjustment and improving the user's experience with the intelligent sensing airbag.

[0185] In an optional embodiment, the specific method by which the first determining module 302 determines the target weights corresponding to the training information may include:

[0186] Obtain the preset weights corresponding to each training sub-information in the training information;

[0187] Obtain the user's training difficulty level, which represents the range of training difficulty that the user can tolerate;

[0188] Based on the training difficulty level, the standard range corresponding to each training sub-information in the training information is determined. The standard range corresponding to the training sub-information is used to represent the range of values ​​that can be used as the standard value corresponding to the training sub-information under the training difficulty level.

[0189] For each type of training sub-information in the training information, determine whether the training sub-information exceeds the standard range corresponding to the training sub-information, and obtain the range judgment result of the training sub-information.

[0190] Based on the range judgment results of all training sub-information in the training information, all training sub-information is divided into a first sub-information set and a second sub-information set. The first sub-information set includes all training sub-information that exceeds the corresponding standard range, and the second sub-information set includes all training sub-information that does not exceed the corresponding standard range.

[0191] Based on predetermined weight adjustment conditions, the preset weights corresponding to each training sub-information in the first sub-information set are adjusted to obtain the first weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the first sub-information set.

[0192] Based on the first weight adjustment result, the preset weights corresponding to each training sub-information in the second sub-information set are adjusted to obtain the second weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the second sub-information set.

[0193] As can be seen, the apparatus described in this optional embodiment can determine the standard range corresponding to each type of training sub-information in the training information based on the acquired training difficulty level, determine whether the training sub-information exceeds the corresponding standard range, obtain a range judgment result, and divide all training sub-information into a first sub-information set and a second sub-information set according to the range judgment result. After adjusting the preset weights corresponding to each training sub-information in the first sub-information set to obtain a first weight adjustment result, the second sub-information set is adjusted according to the first weight adjustment result to obtain a second weight adjustment result. This can improve the accuracy of determining the standard range corresponding to the training sub-information, thereby improving the accuracy of dividing the sub-information set, and further improving the accuracy of determining the sub-weights corresponding to the training sub-information, which is beneficial to improving the accuracy of attitude training plan generation.

[0194] In this optional embodiment, the first determining module 302 is further configured to determine the standard force application method corresponding to each planned training posture in the first posture training plan according to the first posture training plan and the training difficulty level. The standard force application method is matched with the training difficulty level. The standard force application method includes the force application part, the force application degree corresponding to the force application part, and the force application duration corresponding to the force application part. The force application part is used to indicate the body part that the user needs to exert force voluntarily when completing the planned training posture.

[0195] The first determining module 302 is also used to determine the control parameters of the intelligent sensing airbag based on the first posture training plan and the standard force application method corresponding to each planned training posture in the first posture training plan.

[0196] As can be seen, the device described in this optional embodiment can also determine the standard force application method corresponding to each planned training posture based on the generated posture training plan and training difficulty level, and determine the control parameters of the intelligent sensing airbag based on the posture training plan and the corresponding standard force application method. This can improve the accuracy of determining the standard force application method corresponding to the training posture, thereby improving the accuracy of determining the control parameters of the intelligent sensing airbag, and further improving the control accuracy of the intelligent sensing airbag, which is beneficial to improving the user's experience with the intelligent sensing airbag.

[0197] In another optional embodiment, the phase completion information may include the target training completion rate of the current training phase, the actual training completion rate of the current training phase, and the completion difference rate of the current training phase.

[0198] The actual training completion rate in the current training phase can be determined in the following ways:

[0199] Based on the user's corresponding posture recognition information, the current posture difference is determined. The current posture difference is used to represent the degree of difference between the user's current posture and the standard training posture corresponding to the current posture. The standard training posture corresponding to the current posture is one of the standard training postures in the current training stage.

[0200] The actual training completion rate of the current training phase is calculated based on the posture difference rate corresponding to all standard training postures in the current training phase.

[0201] The completion difference in the current training phase can be determined in the following ways:

[0202] The completion difference of the current training stage is calculated based on the target training completion rate and the actual training completion rate.

[0203] As can be seen, the apparatus described in this optional embodiment can determine the current posture difference based on the user's current posture and the standard training posture corresponding to the current posture, and calculate the actual training completion based on the posture difference corresponding to all standard training postures in the current training stage. This can improve the accuracy of determining the actual training completion, thereby improving the accuracy of stage completion information and thus improving the accuracy of posture training plan generation. Furthermore, it can calculate the completion difference based on the target training completion and the actual training completion, which can improve the accuracy of determining the completion difference and thus improve the accuracy of stage completion information.

[0204] In yet another alternative embodiment, the pose recognition information can be determined in the following way:

[0205] Adaptive weighted fusion processing is performed on the real-time user motion sensing data to obtain multi-sensor fusion data. The motion sensing data is obtained based on a sensor group, which includes a range sensor, an angle sensor, and a pressure sensor. The range sensor is used to collect the user's leg and / or hip lift height sensing data, the angle sensor is used to collect the user's leg and / or hip lift angle sensing data, and the pressure sensor is used to collect the pressure value inside the airbag.

[0206] The multi-sensor fusion data is input into a pre-trained convolutional neural network model to obtain the user's first pose recognition information;

[0207] By analyzing the motion image data corresponding to the motion sensing data, the user's second posture recognition information is obtained. The motion image data is obtained based on the camera device corresponding to the intelligent sensing airbag.

[0208] The user's posture recognition information is determined based on the user's first posture recognition information and the user's second posture recognition information.

[0209] As can be seen, the apparatus described in this optional embodiment can perform fusion processing on the collected motion sensing data to obtain multi-sensor fusion data, input the multi-sensor fusion data into a convolutional neural network model to obtain first posture recognition information, and analyze motion image data to obtain second posture recognition information. Based on the first posture recognition information and the second posture recognition information, the user's posture recognition information is determined, which can improve the accuracy of posture recognition information determination, thereby improving the accuracy of posture recognition information and thus improving the accuracy of posture training plan generation.

[0210] In yet another alternative embodiment, such as Figure 5 As shown, the device may further include:

[0211] The acquisition module 304 is used to acquire the user's second user body information after the generation module 303 generates the user's first posture training plan based on the training information and the target weights corresponding to the training information. The user's second user body information is used to represent the user's body information after posture training using intelligent sensing airbags according to the first posture training plan. The user body information includes one or more combinations of user weight, user height, user body fat percentage, spinal status, pelvic status and leg status.

[0212] The second determining module 305 is used to determine the actual degree of physical change of the user based on the first user's physical information and the second user's physical information;

[0213] The judgment module 306 is used to determine whether the user's actual physical changes have reached the expected physical changes corresponding to the user's training objectives, which include one or more combinations of daily exercise, fitness shaping and posture adjustment.

[0214] The statistics module 307 is used to count the duration of the user's physical changes not meeting the expected physical changes corresponding to the user's training objective when the judgment module 306 determines that the user's actual physical changes have not met the expected physical changes. The duration of the physical changes not meeting the expected physical changes is used to indicate the duration of the user's actual physical changes not meeting the expected physical changes.

[0215] The judgment module 306 is also used to determine whether the duration of the body change failure exceeds the failure duration threshold corresponding to the first posture training plan;

[0216] The adjustment module 308 is used to adjust the first posture training plan and obtain the second posture training plan when the judgment module 306 determines that the duration of the body change that does not meet the standard exceeds the threshold of the duration of the failure to meet the standard corresponding to the first posture training plan.

[0217] As can be seen, the device described in this optional embodiment can determine the actual degree of physical change of a user after training according to the posture training plan. When it is determined that the actual degree of physical change of the user has not reached the expected degree of physical change corresponding to the user's training objective, the duration of the unmet physical change is counted. If the duration of the unmet physical change exceeds the unmet duration threshold corresponding to the posture training plan, the posture training plan is adjusted according to the actual degree of physical change and the duration of the unmet physical change. This can improve the accuracy of determining the degree of physical change of the user, thereby improving the accuracy of judging whether the degree of physical change of the user has reached the target, and further improving the accuracy of adjusting the posture training plan. This is beneficial to reducing the possibility of user injury during training and improving the accuracy of user posture adjustment.

[0218] In this optional embodiment, the second determining module 305 is further configured to determine the user's body change trend based on the user's actual body change and the user's historical body change when the determining module 306 determines that the user's actual body change reaches the expected body change corresponding to the user's training objective. The historical body change is the body change determined based on the user's body information before / after using the intelligent sensing airbag for posture training according to the historical posture training plan.

[0219] The judgment module 306 is also used to determine whether the fluctuation amplitude of the body's change trend exceeds the fluctuation amplitude threshold.

[0220] Among them, such as Figure 5 As shown, the device may further include:

[0221] Analysis module 309: When judgment module 306 determines that the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold, it analyzes the first posture training plan and the user's historical posture training plan to obtain the cause of the trend fluctuation. The cause of the trend fluctuation is used to indicate the reason why the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold.

[0222] The adjustment module 308 is also used to adjust the first posture training plan according to the cause of the trend fluctuation to obtain the third posture training plan, so that after the user uses the intelligent sensing airbag for posture training according to the third posture training plan, the fluctuation amplitude of the user's body change trend is lower than the fluctuation amplitude threshold.

[0223] As can be seen, the device described in this optional embodiment can also determine whether the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold when it is determined that the user's actual body change has reached the expected body change corresponding to the user's training purpose. If it exceeds the threshold, the device analyzes the posture training plan and historical posture training plans to obtain the reason for the excessive fluctuation amplitude of the body change trend. The device can then adjust the posture training plan according to the reason for the trend fluctuation, which can improve the accuracy of determining the user's body change trend, thereby improving the flexibility and accuracy of adjusting the posture training plan. This is beneficial to improving the fit between the posture training plan and the user's body change trend, and thus improving the accuracy of adjusting the user's posture.

[0224] Example 4

[0225] Please see Figure 6 , Figure 6 This is a schematic diagram of another posture training plan generation device based on intelligent sensing airbags disclosed in an embodiment of the present invention. Figure 6 As shown, the posture training plan generation device based on intelligent sensing airbags may include:

[0226] Memory 401 storing executable program code;

[0227] Processor 402 coupled to memory 401;

[0228] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the posture training plan generation method based on intelligent sensing airbags described in Embodiment 1 or Embodiment 2 of the present invention.

[0229] Example 5

[0230] This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute the steps in the posture training plan generation method based on intelligent sensing airbags described in Embodiment 1 or Embodiment 2 of this invention.

[0231] Example 6

[0232] This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the posture training plan generation method based on intelligent sensing airbags described in Embodiment 1 or Embodiment 2.

[0233] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0234] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0235] Finally, it should be noted that the posture training plan generation method and apparatus based on intelligent sensing airbags disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for generating attitude training plans based on intelligent sensing airbags, characterized in that, The method is applied to a smart sensing airbag, the smart sensing airbag comprising a sensor array and at least one airbag, the method comprising: Collect training information corresponding to the user. The training information includes one or more combinations of first user body information, posture recognition information and stage completion information. The posture recognition information includes the current posture of the user that has been identified. The stage completion information is used to indicate the training completion degree of the user in the current training stage. Determine the target weights corresponding to the training information, wherein the target weights corresponding to the training information include the sub-weights corresponding to each type of training sub-information in the training information, and the sum of the sub-weights corresponding to all the training sub-information is equal to one; Based on the training information and the target weights corresponding to the training information, a first posture training plan for the user is generated. The first posture training plan includes one or more combinations of planned training posture, training duration, training frequency, and expected degree of body change. The first posture training plan is used to determine the control parameters of the intelligent sensing airbag. The control parameters of the intelligent sensing airbag are used to control the intelligent sensing airbag to perform the inflation / deflation operation corresponding to the control parameters when the user uses the intelligent sensing airbag for posture training, so as to guide the user to adjust the user's posture according to the first posture training plan. Wherein, determining the target weights corresponding to the training information includes: Obtain the preset weights corresponding to each type of training sub-information in the training information; Obtain the user's training difficulty level, which represents the range of training difficulty the user can tolerate; Based on the training difficulty level, a standard range corresponding to each training sub-information in the training information is determined. The standard range corresponding to the training sub-information is used to represent the range of values ​​that can be used as the standard value corresponding to the training sub-information under the training difficulty level. For each type of training sub-information in the training information, determine whether the training sub-information exceeds the standard range corresponding to the training sub-information, and obtain the range judgment result of the training sub-information; Based on the range judgment results of all the training sub-information in the training information, all the training sub-information is divided into a first sub-information set and a second sub-information set. The first sub-information set includes all training sub-information that exceeds the corresponding standard range, and the second sub-information set includes all training sub-information that does not exceed the corresponding standard range. Based on predetermined weight adjustment conditions, the preset weights corresponding to each training sub-information in the first sub-information set are adjusted to obtain a first weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the first sub-information set. Based on the first weight adjustment result, the preset weights corresponding to each training sub-information in the second sub-information set are adjusted to obtain the second weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the second sub-information set.

2. The posture training plan generation method based on intelligent sensing airbags according to claim 1, characterized in that, The method further includes: Based on the first posture training plan and the training difficulty level, the standard force application method corresponding to each planned training posture in the first posture training plan is determined. The standard force application method is matched with the training difficulty level. The standard force application method includes the force application part, the force application degree corresponding to the force application part, and the force application duration corresponding to the force application part. The force application part is used to indicate the body part that the user needs to exert force voluntarily when completing the planned training posture. The control parameters of the intelligent sensing airbag are determined based on the first posture training plan and the standard force application method corresponding to each of the planned training postures in the first posture training plan.

3. The posture training plan generation method based on intelligent sensing airbags according to claim 1, characterized in that, The stage completion information includes the target training completion rate of the current training stage, the actual training completion rate of the current training stage, and the completion difference rate of the current training stage. The actual training completion rate of the current training phase is determined in the following way: Based on the user's posture recognition information, the current posture difference is determined. The current posture difference is used to represent the degree of difference between the user's current posture and the standard training posture corresponding to the current posture. The standard training posture corresponding to the current posture is one of the standard training postures in the current training stage. The actual training completion rate of the current training phase is calculated based on the posture difference degree corresponding to all the standard training postures in the current training phase. The completion difference of the current training phase is determined in the following way: The completion difference of the current training stage is calculated based on the target training completion rate and the actual training completion rate.

4. The posture training plan generation method based on intelligent sensing airbags according to claim 1, characterized in that, The posture recognition information is determined in the following way: The motion sensing data of the user collected in real time is subjected to adaptive weighted fusion processing to obtain multi-sensor fusion data; wherein, the motion sensing data is obtained based on the sensor group, which includes a distance sensor, an angle sensor and a pressure sensor. The distance sensor is used to collect the lifting height sensing data of the user's legs and / or hips, the angle sensor is used to collect the lifting angle sensing data of the user's legs and / or hips, and the pressure sensor is used to collect the pressure value inside the airbag; The multi-sensor fused data is input into a pre-trained convolutional neural network model to obtain the user's first pose recognition information; By analyzing the motion image data corresponding to the motion sensing data, the user's second posture recognition information is obtained. The motion image data is obtained based on the camera device corresponding to the intelligent sensing airbag. The user's posture recognition information is determined based on the user's first posture recognition information and the user's second posture recognition information.

5. The posture training plan generation method based on intelligent sensing airbags according to any one of claims 1-4, characterized in that, After generating the user's first pose training plan based on the training information and the target weights corresponding to the training information, the method further includes: The user's second body information is obtained. The user's second body information is used to represent the user's body information after posture training using the intelligent sensing airbag according to the first posture training plan. The user's body information includes one or more combinations of user weight, user height, user body fat percentage, spinal status, pelvic status and leg status. Based on the first user's physical information and the second user's physical information, determine the actual degree of physical change of the user; Based on the predetermined user training objectives, it is determined whether the actual degree of physical change of the user has reached the expected degree of physical change corresponding to the user training objectives. The user training objectives include one or more combinations of daily exercise, fitness shaping and posture adjustment. When it is determined that the user's actual physical changes have not reached the expected physical changes corresponding to the user's training objectives, the duration of the user's physical changes not meeting the target is counted. The duration of the physical changes not meeting the target is used to indicate the duration during which the user's actual physical changes have not reached the expected physical changes. Determine whether the duration of the body changes that did not meet the standard exceeds the threshold for the duration of non-compliance corresponding to the first posture training plan; When it is determined that the duration of the body change that does not meet the standard exceeds the threshold of the duration of the failure to meet the standard corresponding to the first posture training plan, the first posture training plan is adjusted according to the actual degree of body change and the duration of the body change that does not meet the standard to obtain a second posture training plan.

6. The posture training plan generation method based on intelligent sensing airbags according to claim 5, characterized in that, The method further includes: When it is determined that the user's actual physical change reaches the expected physical change corresponding to the user's training objective, the user's physical change trend is determined based on the user's actual physical change and the user's historical physical change. The historical physical change is the physical change determined based on the user's physical information before / after using the intelligent sensing airbag for posture training according to the historical posture training plan. Determine whether the fluctuation range of the stated body change trend exceeds the fluctuation range threshold; When it is determined that the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold, the first posture training plan and the user's historical posture training plan are analyzed to obtain the cause of the trend fluctuation. The cause of the trend fluctuation is used to indicate the reason why the fluctuation amplitude of the body change trend exceeds the fluctuation amplitude threshold. Based on the cause of the trend fluctuation, the first posture training plan is adjusted to obtain a third posture training plan, so that after the user performs posture training using the intelligent sensing airbag according to the third posture training plan, the fluctuation amplitude of the user's body change trend is lower than the fluctuation amplitude threshold.

7. A posture training plan generation device based on intelligent sensing airbags, characterized in that, The device is applied to a smart sensing airbag, the smart sensing airbag comprising a sensor array and at least one airbag, the device comprising: The acquisition module is used to acquire training information corresponding to the user. The training information includes one or more combinations of first user body information, posture recognition information and stage completion information. The posture recognition information includes the identified current posture of the user, and the stage completion information is used to indicate the training completion degree of the user in the current training stage. The first determining module is used to determine the target weight corresponding to the training information, wherein the target weight corresponding to the training information includes the sub-weight corresponding to each type of training sub-information in the training information, and the sum of the sub-weights corresponding to all the training sub-information is equal to one; A generation module is used to generate a first posture training plan for the user based on the training information and the target weights corresponding to the training information. The first posture training plan includes one or more combinations of planned training posture, training duration, training frequency, and expected degree of body change. The first posture training plan is used to determine the control parameters of the intelligent sensing airbag. The control parameters of the intelligent sensing airbag are used to control the intelligent sensing airbag to perform the inflation / deflation operation corresponding to the control parameters when the user uses the intelligent sensing airbag for posture training, so as to guide the user to adjust the user's posture according to the first posture training plan. The specific method by which the first determining module determines the target weights corresponding to the training information includes: Obtain the preset weights corresponding to each type of training sub-information in the training information; Obtain the user's training difficulty level, which represents the range of training difficulty the user can tolerate; Based on the training difficulty level, a standard range corresponding to each training sub-information in the training information is determined. The standard range corresponding to the training sub-information is used to represent the range of values ​​that can be used as the standard value corresponding to the training sub-information under the training difficulty level. For each type of training sub-information in the training information, determine whether the training sub-information exceeds the standard range corresponding to the training sub-information, and obtain the range judgment result of the training sub-information; Based on the range judgment results of all the training sub-information in the training information, all the training sub-information is divided into a first sub-information set and a second sub-information set. The first sub-information set includes all training sub-information that exceeds the corresponding standard range, and the second sub-information set includes all training sub-information that does not exceed the corresponding standard range. Based on predetermined weight adjustment conditions, the preset weights corresponding to each training sub-information in the first sub-information set are adjusted to obtain a first weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the first sub-information set. Based on the first weight adjustment result, the preset weights corresponding to each training sub-information in the second sub-information set are adjusted to obtain the second weight adjustment result, which includes the sub-weights corresponding to each training sub-information in the second sub-information set.

8. A posture training plan generation device based on intelligent sensing airbags, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the posture training plan generation method based on intelligent sensing airbags as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the posture training plan generation method based on intelligent sensing airbags as described in any one of claims 1-6.

Images