A smart sofa control method and system based on posture recognition

Abstract

The present invention provides a method and system for controlling an intelligent sofa based on posture recognition, which relates to the technical field of intelligent home control. The method comprises: obtaining a human body image on an intelligent sofa; extracting a plurality of human skeleton key points in the human body image; determining the angles of a plurality of human key joints according to each of the human skeleton key points; determining the posture of the human body on the intelligent sofa according to the angles of each of the human key joints; determining a control mode corresponding to the posture according to the posture of the human body on the intelligent sofa; and controlling the posture of the human body on the intelligent sofa by using PI. <supgt; l < / supgt;D<supgt; m

Description

Technical Field

[0001] The present invention relates to the field of smart home control technology, and in particular to a smart sofa control method and system based on posture recognition. Background Art

[0002] As an important part of the home environment, the improvement of the function and comfort of smart sofas directly affects the user's living experience. Users expect to achieve better experience and health support through smart sofas, such as automatically adjusting the sofa copy angle to adapt to different sitting postures, thereby reducing health problems caused by long-term sitting.

[0003] In the existing technical solutions, traditional smart sofas mainly obtain human pressure data through pressure sensors to estimate human posture, resulting in inaccurate adjustment of the backrest angle of the smart sofa, thus affecting the actual user experience.

[0004] In addition, traditional smart sofas are unable to perform real-time data processing and posture recognition, cannot accurately identify complex human postures, have too long a response time, and cannot adjust the backrest angle of the smart sofa in time according to the user's new posture, failing to achieve a practical level of home comfort and intelligence. Summary of the invention

[0005] In order to solve the technical problems that traditional smart sofas mainly obtain human pressure data through pressure sensors to estimate human posture, resulting in inaccurate adjustment of the backrest angle of the smart sofa, thereby affecting the actual user experience, and cannot perform real-time data processing and posture recognition, cannot accurately recognize complex human postures, and have too long a response time. The backrest angle of the smart sofa cannot be adjusted in time according to the user's new posture, and cannot achieve a practical level of home comfort and intelligence, the present invention provides a smart sofa control method and system based on posture recognition.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] First aspect:

[0008] An embodiment of the present invention provides a method for controlling a smart sofa based on posture recognition, comprising:

[0009] S1: Acquire the human body image on the smart sofa;

[0010] S2: extracting a plurality of human skeleton key points in the human body image;

[0011] S3: determining angles of multiple key joints of the human body according to each of the key points of the human skeleton;

[0012] S4: determining the posture of the human body on the smart sofa according to the angles of each of the key joints of the human body;

[0013] S5: determining a control mode corresponding to the posture according to the posture of the human body on the smart sofa;

[0014] S6: Via PI λ D μ The controller controls the smart sofa according to the determined control mode.

[0015] Second aspect:

[0016] An embodiment of the present invention provides a smart sofa control system based on posture recognition, comprising: a memory and one or more processors;

[0017] One or more application programs are stored in the memory, and the one or more application programs are suitable for being executed by the one or more processors to implement the above-mentioned smart sofa control method based on gesture recognition.

[0018] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0019] In the present invention, multiple human skeleton key points in a human body image are extracted, and the angles of multiple human key joints are determined according to each human skeleton key point, and data processing and posture recognition are performed in real time. The posture of a human body on a smart sofa is determined according to the angles of each human key joint. Complex human postures can be accurately recognized without obtaining human pressure data through a pressure sensor, so that the backrest angle adjustment of the smart sofa is more accurate, and the actual user experience is improved. According to the posture of the human body on the smart sofa, a control mode corresponding to the posture is determined, and the PI λ D μ The controller controls the smart sofa according to the determined control mode, has a short response time, and can adjust the backrest angle of the smart sofa in time according to the user's new posture, achieving a practical level of home comfort and intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A flowchart of a method for controlling a smart sofa based on gesture recognition provided by an embodiment of the present invention;

[0022] Figure 2 A schematic structural diagram of an intelligent sofa control system based on posture recognition provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0024] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0025] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0026] refer to Figure 1 , which shows a flow chart of a smart sofa control method based on posture recognition provided by an embodiment of the present invention.

[0027] The embodiment of the present invention provides a method for controlling a smart sofa based on gesture recognition, which can be implemented by a smart sofa control device based on gesture recognition, and the smart sofa control device based on gesture recognition can be a terminal or a server. The processing flow of the method for controlling a smart sofa based on gesture recognition can include the following steps:

[0028] S1: Acquire a human body image on the smart sofa.

[0029] S2: Extract multiple human skeleton key points in human images.

[0030] In a possible implementation manner, S1 specifically includes:

[0031] The human body image on the smart sofa is obtained through the Kinect2.0 sensor.

[0032] S2 is specifically:

[0033] Through the Kinect2.0 sensor, multiple human skeleton key points in the human body image are extracted.

[0034] It should be noted that the Kinect2.0 sensor is a high-performance motion capture and depth sensing device launched by Microsoft, mainly used for Xbox One game consoles and Windows platforms. It can capture and analyze user movements, sounds and environmental information in real time through multiple sensors and advanced computer vision technology. Its main components include RGB cameras, infrared sensors and microphone arrays. The RGB camera is used to capture color images, while the infrared sensor can work under different lighting conditions to capture depth information and then build models of objects and characters in three-dimensional space. The microphone array is used for sound source localization and speech recognition, allowing the device to understand the user's voice commands. A notable feature of Kinect2.0 is its high-precision skeletal tracking function, which can identify 25 key points of the human skeleton, allowing it to track in real time during the interaction of multiple users.

[0035] Among them, the key points of the human skeleton include: head, neck, back shoulder, left shoulder, right shoulder, middle of the back, left elbow, right elbow, bottom of the back, left hip, right hip, left wrist, right wrist, left hand, right hand, left thumb, right thumb, left hand tip, right hand tip, left knee, right knee, left ankle, right ankle, left foot and right foot.

[0036] In the present invention, Kinect2.0 can track and identify 25 key bone points of the human body. This high-precision bone tracking function enables the smart sofa to capture the user's posture and movement very accurately, thereby accurately analyzing the user's posture. This is very critical to ensure that the reaction of the smart sofa is highly synchronized with the user's movements. Kinect2.0 integrates RGB cameras and infrared sensors, allowing the device to work effectively in various lighting environments. Infrared sensors are particularly suitable for capturing depth information in low-light or dark environments, which ensures that the smart sofa can accurately detect and respond to the user's posture under different environmental conditions.

[0037] S3: Determine the angles of multiple key joints of the human body based on the key points of each human skeleton.

[0038] Optionally, the key joints of the human body include: neck joints (head, neck and back shoulders), left shoulder joints (neck, left shoulder and left elbow), right shoulder joints (neck, right shoulder and right elbow), left elbow joints (left shoulder, left elbow and left wrist), right elbow joints (right shoulder, right elbow and right wrist), upper spine joints (back shoulders, middle back and bottom back), left hip joints (bottom back, left hip and left knee), right hip joints (bottom back, right hip and right knee), left knee joints (left hip, left knee and left ankle), right knee joints (right hip, right knee and right ankle), left ankle joints (left knee, left ankle and left foot) and right ankle joints (right knee, right ankle and right foot).

[0039] In the present invention, accurate measurement of the angles of multiple key human joints can more accurately identify human postures. According to the angles of key human joints, the smart sofa can automatically adjust the backrest angle, seat cushion hardness, etc. to adapt to the body structure and preferences of different users and provide a personalized comfort experience. By monitoring joint angles, the smart sofa can identify postures that may cause discomfort or long-term health problems, and make timely adjustments to promote a healthier sitting posture, which helps reduce stress and fatigue.

[0040] In a possible implementation, S3 specifically includes sub-steps S301 and S302:

[0041] S301: Calculate the displacement vectors between key points of human skeletons.

[0042] Optionally, the displacement vector between the key points of each human skeleton is calculated according to the following formula:

[0043]

[0044] in, Represents the key points of the human skeleton A To the key points of the human skeleton O The displacement vector, x 0 represents the key points of human skeleton O The horizontal axis of x 1 represents the key point of human skeleton A The horizontal axis of y 0 represents the key points of human skeleton O The vertical coordinate of y 1 represents the key point of human skeleton A The vertical coordinate of Represents the key points of the human skeleton A To the key points of the human skeleton B The displacement vector, x 2 represents the key points of the human skeleton B The horizontal axis of y 2 represents the key points of the human skeleton B The vertical coordinate of .

[0045] S302: Determine angles of multiple key joints of the human body according to the displacement vector.

[0046] Optionally, the angles of multiple key human joints are determined according to the following formula:

[0047]

[0048] in, i Indicates the angles of key joints of the human body, represents the arccosine function, Represents the magnitude of the displacement vector.

[0049] It should be noted that the key points of the human skeleton A It represents the key points of the human skeleton at the key joints of the human body. B It represents the upstream human skeleton key points of the key joints of the human body. O It represents the key points of the human skeleton downstream of the key joints of the human body.

[0050] For example, when calculating the angle of the neck joint, the key points of the human skeleton A For the neck, key points of the human skeleton B For the head, the key points of the human skeleton O For the back shoulders.

[0051] In the present invention, by calculating the displacement vector and using geometric methods (such as the arccosine function) to calculate the angle, the specific angle of the joint can be obtained very accurately. The angle data of each joint can help the system understand the user's physical state and dynamics more comprehensively, such as sitting posture, bending over, etc. This detailed capture is very valuable for identifying the user's behavior patterns, predicting their needs, and adjusting the sofa settings accordingly. By accurately monitoring the joint angles, the smart sofa can help users maintain a healthy sitting posture and avoid physical pain or long-term injuries that may be caused by maintaining a bad posture for a long time. The system can prompt the user to make necessary body adjustments or stretches to promote blood circulation and muscle relaxation.

[0052] S4: Determine the posture of the human body on the smart sofa according to the angles of each key joint of the human body.

[0053] In the present invention, by monitoring and analyzing joint angles, the smart sofa can identify the user's specific posture, thereby providing personalized support and comfort adjustment for each posture. This personalized response not only improves the user's comfort experience, but also meets the specific needs of different users. Accurate posture recognition allows the smart sofa to detect potential bad sitting postures that may cause physical pain or other health problems. The sofa can automatically adjust to prompt the user to change the sitting posture, which helps prevent muscle tension and fatigue caused by maintaining the same posture for a long time.

[0054] In a possible implementation, S4 specifically includes sub-steps S401 and S402:

[0055] S401: Calculate the matching scores between the angles of each key joint of the human body and various postures:

[0056]

[0057] in, Indicates j The matching scores of the poses,N represents the total number of key joints in the human body, Indicates The weight coefficients of the angles of the key joints of the human body, Indicates The angles of the key joints of the human body, Indicates The angles of the key joints of the human body are j Preset angle range for different postures, Indicates The angles of the key joints of the human body are j The minimum preset angle range under various postures, Indicates The angles of the key joints of the human body are j The maximum preset angle range in various postures, Represents a conditional function.

[0058] It should be noted that the matching score refers to a quantitative score that calculates the similarity between the user's current posture and a set of preset postures. This score is determined by comparing whether the angles of each key joint of the human body match the range of the corresponding joint angles in the preset posture. If the angle of a joint is within its corresponding preset range, a positive value is added to the matching score of the posture. If it is not within the range, the joint does not contribute to the matching score. By summarizing the scores of all key joints, a total matching score is finally obtained, which is used to identify and select the best matching posture.

[0059] It should be noted that those skilled in the art can set the size of the preset angle range according to actual needs, and the present invention is not limited thereto.

[0060] S402: Determine the posture with the highest matching score as the posture of the human body on the smart sofa.

[0061] Optionally, the posture includes a reading posture, a resting posture and a working posture.

[0062] In the present invention, the use of a matching score mechanism allows the smart sofa to identify the user's current posture with high accuracy. This method can accurately determine the user's closest posture type, such as reading, resting or working posture, by comparing the degree of matching of each joint angle with the preset angle range in detail. By calculating the matching score in real time, the smart sofa can dynamically adjust its settings to adapt to the user's most likely posture, providing more appropriate support and comfort. For example, if the user is detected to be in a reading posture, the sofa can automatically adjust the backrest angle to optimize the reading experience. By defining preset angle ranges for multiple postures, the smart sofa can adapt to a variety of life and work scenarios, so that the same furniture product can provide optimal support in different environments and needs.

[0063] S5: According to the posture of the human body on the smart sofa, determine a control mode corresponding to the posture.

[0064] Optionally, the control modes include: a reading mode (when the human body is in a reading posture), a resting mode (when the human body is in a resting posture) and a working mode (when the human body is in a working posture).

[0065] In the present invention, the smart sofa can provide the most suitable support and comfort by recognizing the user's posture and adjusting the sofa to the corresponding control mode (such as reading, resting, working). For example, the reading mode may require a straighter backrest and moderate seat cushion hardness to support long-term sitting and stable vision. Different control modes adapt to different activity needs from reading, resting to working, making the sofa not only a seat, but a multifunctional living and working space. Such functions are particularly suitable for the diversity of modern home life and work.

[0066] S6: Via PI λ D μ The controller controls the smart sofa according to the determined control mode.

[0067] It should be noted that PI λ D μ The controller is an advanced controller design that uses the concept of fractional calculus to expand the traditional PID controller and add more flexible adjustment functions. In the traditional PID (proportional-integral-differential) controller, the three elements of proportional, integral, and differential control the output of the system to achieve the desired system performance. These control elements are achieved by adjusting the proportional gain coefficient ( K p )、Integral gain coefficient( K i ) and the differential gain coefficient ( K d ). By introducing fractional integration (integral order l ) and fractional differentials (differentiation order m ), allowing the controller to more finely tune its response to errors. The concept of fractional order comes from fractional calculus, which expands the definition of integration and differentiation, and is not limited to integer orders, but can use real orders. Such a controller provides additional degrees of freedom, allowing the controller to be more flexible to adapt to different types of dynamic systems, improving the stability and response speed of the system.

[0068] In the present invention, PI λ D μThe controller provides a more precise and sensitive control effect than the traditional PID controller by introducing fractional-order integration and differentiation. This means that the response of the sofa can more accurately match the user's movements and posture changes, thereby providing higher comfort and satisfaction. λ D μ The controller can effectively reduce overshoot and oscillation in the system response, which is particularly important for the user experience. When adjusting the backrest angle or seat cushion hardness, the set state can be reached more smoothly, avoiding discomfort caused by rapid changes.

[0069] In a possible implementation, S6 specifically includes sub-steps S601 to S603:

[0070] S601: Obtain the target backrest angle of the smart sofa in the current control mode.

[0071] S602: Acquire the actual backrest angle of the smart sofa in the current control mode.

[0072] S603: Through PI λ D μ The controller outputs control signals to control the smart sofa.

[0073] Optionally, according to the following formula, through PI λ D μ Controller, output control signal, controls the smart sofa:

[0074]

[0075] in, u ( t )express t The control signal at the moment, K p represents the proportional gain coefficient, e ( t )express t The deviation between the target backrest angle and the actual backrest angle at the moment, K i represents the integral gain coefficient, K d represents the differential gain coefficient, D -λ represents fractional-order integral operation, l represents the integration order, D μ represents fractional differential operation, m represents the differential order, L ( t )express t The actual backrest angle of the smart sofa at the moment, L0 represents the target backrest angle of the smart sofa.

[0076] Optionally, for a smart sofa, during the control mode switching process, the control quantity that needs to be changed is not only the backrest angle, but also may include: the seat cushion hardness, backrest hardness and headrest angle of the smart sofa. For the control of the seat cushion hardness, backrest hardness and headrest angle, reference may be made to the control scheme for the backrest angle described above, and the present invention will not elaborate on this in order to avoid repetition.

[0077] In the present invention, by introducing the integration order l and the differential order m , P.I. λ D μ The controller can adjust the control strategy more finely to accurately match the dynamic characteristics of the smart sofa. Fractional-order control elements allow the controller to have better adjustment capabilities, so that the system can quickly and stably reach the target state, reduce overshoot and oscillation, and provide a smoother user experience. Due to its additional adjustment freedom, the fractional-order controller can better adapt to various types of dynamic loads and changes in external conditions, thereby maintaining excellent performance in different usage situations.

[0078] In a possible implementation, the smart sofa control method based on gesture recognition further includes:

[0079] S7: With the goal of improving the comfort index during the control mode switching process and reducing the total time spent in the control mode switching process, the PI λ D μ Optimize the controller.

[0080] It should be noted that the cuckoo search algorithm is a heuristic optimization algorithm inspired by the parasitic reproduction behavior of cuckoos, especially their habit of parasitizing the nests of other birds. In this algorithm, each nest represents a potential solution, and the eggs in the nest symbolize the various parameters of the solution. When the algorithm starts, a series of nests are randomly generated as the initial solution. The fitness of each nest is then evaluated, and the higher the fitness, the closer the solution is to the optimal. The cuckoo will choose a nest to lay eggs, that is, generate a new potential solution, and use the long-distance random steps of Levy flight to help jump out of the local optimal solution. If the newly laid eggs are not found by the parasite host and the solution is better than the solution in the current nest, it will be replaced. The whole process is repeated, and the nest with the highest fitness is retained until the predetermined termination condition is reached, such as the maximum number of iterations or a sufficiently good solution quality is achieved. The cuckoo search algorithm is widely used in parameter optimization problems in engineering design, resource management and scientific research for its simplicity and efficiency. It can effectively explore and utilize the search space and avoid falling into the local optimum too early.

[0081] Furthermore, the cuckoo search algorithm is improved by adaptively adjusting the step size factor, discovery probability and scaling factor according to the current number of iterations of the algorithm, so that the algorithm can dynamically adjust the search strategy according to the obtained results during the search process, thereby more efficiently avoiding local optimal solutions and accelerating the search for the global optimal solution.

[0082] In the present invention, the improved version of the cuckoo search algorithm improves the ability of the algorithm to jump out of the local optimal solution and accelerates the convergence to the global optimal solution by dynamically adjusting the search strategy (such as step size factor, discovery probability and scaling factor), thereby improving the quality and speed of optimization. The cuckoo search algorithm can automatically adjust the search parameters according to the information obtained during the iteration process, which makes the algorithm more flexible and adaptable, and can cope with complex and changeable optimization problems, such as multi-parameter adjustment of the intelligent sofa control system.

[0083] In a possible implementation, S7 specifically includes sub-steps S701 and S702:

[0084] S701: Construct an objective function with the goal of improving the comfort index during the control mode switching process and reducing the total time spent in the control mode switching process.

[0085] In a possible implementation, the objective function is specifically:

[0086]

[0087] in, represents the objective function, max represents the maximum value, Indicates PI λ D μ The set of control parameters of the controller, K Indicates the total time spent in the control mode switching process. r ( k )express k Comfort index at all times, or Represents the weight coefficient of the total duration, where PI λ D μ The control parameters of the controller include proportional gain coefficient, integral gain coefficient, differential gain coefficient, integral order and differential order.

[0088] S702: According to the objective function, PI is searched by using the cuckoo search algorithm. λ D μ Optimize the controller.

[0089] In the present invention, by setting the objective function to maximize the comfort index during the control mode switching process, the optimization process is ensured to focus on improving the user experience. Such an objective function not only considers the optimal value of the control parameter, but also incorporates time efficiency, that is, seeking a control solution that achieves the highest comfort in the shortest time. The cuckoo search algorithm is used to optimize PI λ D μ The parameters of the controller are calculated by this algorithm, which is well-known for its efficient global search capability. The cuckoo search algorithm can effectively avoid falling into the local optimal solution and is more likely to find the global optimal solution, which is especially important for the parameter adjustment of complex control systems.

[0090] In a possible implementation manner, the calculation method of the comfort index specifically includes:

[0091] By evenly setting multiple pressure sensors under the seat cushion and backrest of the smart sofa, the human body pressure data on the smart sofa is obtained:

[0092]

[0093] in, P i Indicates i The human body pressure data obtained by the pressure sensor, F i Indicates that the human body acts on i The pressure on the pressure sensor, i= 1,2,… n , n Indicates the total number of pressure sensors, Indicates the sensing area of ​​the pressure sensor.

[0094] According to the human body pressure data obtained by each pressure sensor, the maximum human body pressure data is determined:

[0095]

[0096] in, P max Indicates the maximum human body pressure data, max means taking the maximum value, P 1 indicates the human body pressure data obtained by the first pressure sensor. P 2 represents the human body pressure data obtained by the second pressure sensor. P n Indicates n Human body pressure data obtained by a pressure sensor.

[0097] According to the human body pressure data obtained by each pressure sensor, the average human body pressure data is determined:

[0098]

[0099] in, P avg Indicates average human body pressure data.

[0100] According to the human body pressure data obtained by each pressure sensor, the asymmetry coefficient of the left and right sides of the smart sofa is determined:

[0101]

[0102] in, CV represents the asymmetry coefficient, P il Indicates the left side of the smart sofa i The human body pressure data obtained by the pressure sensor, P ir Indicates the right side of the smart sofa i Human body pressure data obtained by a pressure sensor.

[0103] According to the human body pressure data obtained by each pressure sensor, the maximum pressure gradient of the human body pressure distribution on the smart sofa is determined:

[0104]

[0105] in, g max represents the maximum pressure gradient, Indicates that the human body pressure data obtained by the pressure sensor is x The rate of change in direction, Indicates that the human body pressure data obtained by the pressure sensor is y The rate of change in direction.

[0106] According to the human body pressure data obtained by each pressure sensor, the average pressure gradient of the human body pressure distribution on the smart sofa is determined:

[0107]

[0108] in, g avg represents the mean pressure gradient, Indicates i The human body pressure data obtained by the pressure sensor is x The rate of change in direction, No. i The human body pressure data obtained by the pressure sensor is y The rate of change in direction.

[0109] The comfort index is calculated based on the maximum human body pressure data, average human body pressure data, asymmetry coefficient, maximum pressure gradient and average pressure gradient:

[0110]

[0111] in, r represents the comfort index, oh 1 represents the weight coefficient of the maximum human body pressure data, oh 2 represents the weight coefficient of average human body pressure data, oh 3 represents the weight coefficient of the asymmetric coefficient, oh 4 represents the weight coefficient of the maximum pressure gradient, oh 5 represents the weight coefficient of the average pressure gradient.

[0112] Among them, the maximum human body pressure data refers to the maximum pressure value detected on a certain pressure sensor on the smart sofa, that is, the maximum pressure exerted by the human body on a certain point of the sofa. It reflects the maximum pressure on a certain part of the human body and affects the overall comfort.

[0113] The average human body pressure data refers to the average value of the pressure data obtained by all pressure sensors. This parameter is used to evaluate the overall pressure distribution of the human body on the sofa and reflects the uniformity of the pressure.

[0114] The asymmetry coefficient refers to the pressure difference between the left and right sides of the smart sofa. By comparing the pressure data on the left and right sides of the sofa, it measures whether the human body is leaning to one side on the sofa, reflecting the balance of the posture.

[0115] The maximum pressure gradient refers to the maximum rate of change of human body pressure data in different directions (x or y) on the sofa surface. It indicates the degree of change in pressure distribution and reflects the stress conditions in certain local areas.

[0116] The average pressure gradient refers to the average value of the pressure change rate obtained by all pressure sensors, which indicates the uniformity of pressure distribution of the human body on the sofa.

[0117] Among them, the comfort index is an important basis for measuring the comfort of smart sofas. The higher the value, the better the user experience.

[0118] In the present invention, the pressure sensor array can be used to monitor and analyze the human body pressure distribution on the sofa in detail. This detailed data collection provides a reliable basis for evaluating comfort, ensuring that the weight distribution and pressure points of each user can be accurately captured. By analyzing the maximum pressure data, average pressure data, asymmetry coefficient and pressure gradient, designers can better understand how to optimize the structure and materials of the sofa, such as adjusting the hardness of the cushion, support structure and backrest design to improve the user's comfort and support. By calculating the comfort index and feeding it back to the control system of the smart sofa in real time, the sofa can automatically adjust according to the user's actual usage to provide the best support and comfort, thereby greatly improving the user's satisfaction and experience quality.

[0119] Optionally, S702 specifically includes sub-steps S7021 to S7028:

[0120] S7021: Initialize parameters and set the maximum number of iterations of the cuckoo search algorithm.

[0121] S7022: Randomly generate an initial population, which includes multiple bird nests, each of which represents a feasible PI λ D μ A collection of control parameters for the controller.

[0122] S7023: Using the objective function as the fitness function, calculating the fitness value of each bird's nest, and taking the bird's nest with the maximum fitness value as the current bird's nest.

[0123] S7024: Adaptively adjust the step size factor, discovery probability, and scaling factor according to the current number of iterations:

[0124]

[0125] in, α t Indicates t The step size factor for the iteration, α max represents the maximum value of the step factor, T represents the maximum number of iterations, α min represents the minimum value of the step factor, Indicates t The first iteration i The probability of finding a bird's nest, P max represents the maximum value of the discovery probability, P min represents the minimum value of the discovery probability, Indicates t The first iteration i The fitness value of a bird's nest, Indicates t The maximum fitness value at the iteration, No. t The first iteration i The scaling factor of the nests, represents the maximum value of the scaling factor, represents the minimum value of the scaling factor, Indicates t The minimum fitness value at the iteration.

[0126] S7025: Update the current bird's nest through Levy flight operation according to the adaptively adjusted step size factor:

[0127]

[0128] in, Indicates t+ The first iteration i A bird's nest, Indicates t The first iteration i A bird's nest, α represents the step size factor after adaptive adjustment, represents the dot multiplication operation, levy ( β ) represents the random step size generated by the Levy flight operation.

[0129] It should be noted that the Levy flight operation is a random step used in optimization algorithms, characterized by its step size following the Levy distribution, which allows the algorithm to perform long-distance exploration. Compared with standard random walks, Levy flights can cover a wide search space more effectively, help the algorithm avoid falling into local optimal solutions, and speed up the process of finding the global optimal solution. This method is particularly useful when solving complex optimization problems because it can explore farther areas, not just the local neighborhood.

[0130] S7026: Calculate the fitness value of the updated bird's nest, and compare the fitness value of the updated bird's nest with the fitness value of the current bird's nest. When the fitness value of the updated bird's nest is greater than the fitness value of the current bird's nest, replace the current bird's nest with the updated bird's nest. When the fitness value of the updated bird's nest is less than or equal to the fitness value of the current bird's nest, keep the current bird's nest unchanged.

[0131] S7027: Generate a random number in the range of 0 to 1, and compare the random number with the adaptively adjusted discovery probability. When the random number is greater than the adaptively adjusted discovery probability, the current bird nest is discovered, and the current bird nest is updated according to the adaptively adjusted scaling factor. When the random number is less than or equal to the adaptively adjusted discovery probability, the current bird nest is not discovered, and the current bird nest remains unchanged.

[0132] Among them, when the random number in S7027 is greater than the adaptively adjusted discovery probability, the current bird nest is discovered, and the current bird nest is updated according to the adaptively adjusted scaling factor, specifically:

[0133]

[0134] in, Indicates the number of times the current bird's nest is discovered. t+ The first iteration i A bird's nest, Indicates the number of times the current bird's nest is discovered. t The first iteration i A bird's nest, c Represents the adaptively adjusted scaling factor, and Indicates the number of times the current bird's nest is discovered. t Two individuals are randomly selected from the population at the iteration.

[0135] S7028: Repeat sub-steps S7025 to S7027 until the maximum number of iterations is reached.

[0136] In the present invention, the PI can be searched by using the cuckoo search algorithm. λ D μ The multiple parameters of the controller are comprehensively optimized to ensure that the performance of the controller is optimal in actual operation. This algorithm is particularly suitable for dealing with optimization problems with multiple variables and complex objective functions. By adaptively adjusting the step size factor, discovery probability, and scaling factor, the optimization algorithm can flexibly adjust the search strategy according to the actual situation during the search process. This not only improves the search efficiency, but also increases the possibility of finding the global optimal solution. The cuckoo search algorithm uses the characteristics of Levy flight and is able to perform long-distance exploration steps, which helps the algorithm jump out of the local optimum and speed up the search for the global optimal solution.

[0137] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0138] In the present invention, multiple human skeleton key points in a human body image are extracted, and the angles of multiple human key joints are determined according to each human skeleton key point, and data processing and posture recognition are performed in real time. The posture of a human body on a smart sofa is determined according to the angles of each human key joint. Complex human postures can be accurately recognized without obtaining human pressure data through a pressure sensor, so that the backrest angle adjustment of the smart sofa is more accurate, and the actual user experience is improved. According to the posture of the human body on the smart sofa, a control mode corresponding to the posture is determined, and the PI λ D μ The controller controls the smart sofa according to the determined control mode, has a short response time, and can adjust the backrest angle of the smart sofa in time according to the user's new posture, achieving a practical level of home comfort and intelligence.

[0139] refer to Figure 2 , showing a structural schematic diagram of an intelligent sofa control system based on posture recognition provided by the present invention.

[0140] The present invention further provides a smart sofa control system 30 based on gesture recognition, comprising: a memory 303 and one or more processors 301 .

[0141] One or more application programs are stored in the memory 303 , and the one or more application programs are suitable for being executed by the one or more processors 301 to implement the smart sofa control method based on gesture recognition described in the method embodiment.

[0142] The intelligent sofa control system 30 based on gesture recognition includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, through a bus 302.

[0143] The structure of the smart sofa control system 30 based on gesture recognition does not constitute a limitation to the embodiment of the present invention.

[0144] Processor 301 may be a CPU, a general purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. Processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0145] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI bus or an EISA bus, etc. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0146] The memory 303 can be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or an EEPROM, a CD-ROM or other optical disk storage, an optical disk storage (including a compressed optical disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0147] It should be noted that the smart sofa control system 30 based on gesture recognition can implement the above-mentioned smart sofa control method based on gesture recognition, and can achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0148] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0149] In the present invention, multiple human skeleton key points in a human body image are extracted, and the angles of multiple human key joints are determined according to each human skeleton key point, and data processing and posture recognition are performed in real time. The posture of a human body on a smart sofa is determined according to the angles of each human key joint. Complex human postures can be accurately recognized without obtaining human pressure data through a pressure sensor, so that the backrest angle adjustment of the smart sofa is more accurate, and the actual user experience is improved. According to the posture of the human body on the smart sofa, a control mode corresponding to the posture is determined, and the PI λ D μ The controller controls the smart sofa according to the determined control mode, has a short response time, and can adjust the backrest angle of the smart sofa in time according to the user's new posture, achieving a practical level of home comfort and intelligence.

[0150] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0151] There are a few points to note:

[0152] (1) The drawings of the embodiments of the present invention only involve structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0153] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0154] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0155] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A smart sofa control method based on posture recognition, characterized in that: include: S1: Acquire the human body image on the smart sofa; S2: extracting a plurality of human skeleton key points in the human body image; S3: determining angles of multiple key joints of the human body according to each of the key points of the human skeleton; S4: determining the posture of the human body on the smart sofa according to the angles of each of the key joints of the human body; S5: determining a control mode corresponding to the posture according to the posture of the human body on the smart sofa; S6: Via PI λ D μ A controller, controlling the smart sofa according to a determined control mode; S7: With the goal of improving the comfort index during the control mode switching process and reducing the total time spent in the control mode switching process, the PI is searched by the cuckoo search algorithm. λ D μ Optimize the controller; Wherein, the S7 specifically includes: S701: constructing an objective function with the goal of improving the comfort index during the control mode switching process and reducing the total time spent in the control mode switching process; S702: According to the objective function, the PI is searched by using a cuckoo search algorithm. λ D μ Optimize the controller; Wherein, the objective function is specifically: in, represents the objective function, max represents the maximum value, Indicates PI λ D μ The set of control parameters of the controller, K Indicates the total time spent in the control mode switching process. ρ ( k )express k Comfort index at all times, η Represents the weight coefficient of the total duration, where the PI λ D μ The control parameters of the controller include proportional gain coefficient, integral gain coefficient, differential gain coefficient, integral order and differential order; The calculation method of the comfort index specifically includes: By evenly arranging a plurality of pressure sensors under the seat cushion and the backrest of the smart sofa, the pressure data of the human body on the smart sofa is obtained: in, P i Indicates i The human body pressure data obtained by the pressure sensor, F i Indicates that the human body acts on i The pressure on the pressure sensor, i= 1,2,… n , n Indicates the total number of pressure sensors, Indicates the sensing area of ​​the pressure sensor; According to the human body pressure data acquired by each of the pressure sensors, the maximum human body pressure data is determined: in, P max Indicates the maximum human body pressure data, max means taking the maximum value, P 1 indicates the human body pressure data obtained by the first pressure sensor. P 2 represents the human body pressure data obtained by the second pressure sensor. P n Indicates n Human body pressure data obtained by a pressure sensor; Determine average human body pressure data according to the human body pressure data acquired by each of the pressure sensors: in, P avg Indicates average human body pressure data; According to the human body pressure data acquired by each of the pressure sensors, the asymmetry coefficients of the left and right sides of the smart sofa are determined: in, CV represents the asymmetry coefficient, P il Indicates the left side of the smart sofa i The human body pressure data obtained by the pressure sensor, P ir Indicates the right side of the smart sofa i Human body pressure data obtained by a pressure sensor; According to the human body pressure data acquired by each of the pressure sensors, the maximum pressure gradient of the human body pressure distribution on the smart sofa is determined: in, g max represents the maximum pressure gradient, Indicates that the human body pressure data obtained by the pressure sensor is x The rate of change in direction, Indicates that the human body pressure data obtained by the pressure sensor is y rate of change in direction; According to the human body pressure data acquired by each of the pressure sensors, the average pressure gradient of the human body pressure distribution on the smart sofa is determined: in, g avg represents the mean pressure gradient, Indicates i The human body pressure data obtained by the pressure sensor is x The rate of change in direction, No. i The human body pressure data obtained by the pressure sensor is y rate of change in direction; The comfort index is calculated according to the maximum human body pressure data, the average human body pressure data, the asymmetry coefficient, the maximum pressure gradient and the average pressure gradient: in, ρ represents the comfort index, ω 1 represents the weight coefficient of the maximum human body pressure data, ω 2 represents the weight coefficient of average human body pressure data, ω 3 represents the weight coefficient of the asymmetric coefficient, ω 4 represents the weight coefficient of the maximum pressure gradient, ω 5 represents the weight coefficient of the average pressure gradient.

2. The intelligent sofa control method based on posture recognition according to claim 1, characterized in that: The S1 is specifically: The human body image on the smart sofa is obtained through the Kinect2.0 sensor; The S2 is specifically: A plurality of human skeleton key points in the human body image are extracted through a Kinect 2.0 sensor.

3. The intelligent sofa control method based on posture recognition according to claim 1, characterized in that: The S3 specifically includes: S301: Calculate the displacement vector between each of the key points of the human skeleton; S302: Determine angles of multiple key joints of the human body according to the displacement vector.

4. The intelligent sofa control method based on posture recognition according to claim 1, characterized in that: The S4 specifically includes: S401: Calculate the matching scores between the angles of each of the key joints of the human body and various postures: in, Indicates j The matching scores of the poses, N represents the total number of key joints in the human body, Indicates The weight coefficients of the angles of the key joints of the human body, Indicates The angles of the key joints of the human body, Indicates The angles of the key joints of the human body are j Preset angle range for different postures, Indicates The angles of the key joints of the human body are j The minimum preset angle range under various postures, Indicates The angles of the key joints of the human body are j The maximum preset angle range in various postures, represents the conditional function; S402: Determine the posture with the highest matching score as the posture of the human body on the smart sofa.

5. The intelligent sofa control method based on posture recognition according to claim 1, characterized in that: The S6 specifically includes: S601: Obtaining a target backrest angle of the smart sofa in a current control mode; S602: Acquire the actual backrest angle of the smart sofa in the current control mode; S603: Through the PI λ D μ The controller outputs a control signal to control the smart sofa.

6. An intelligent sofa control system based on posture recognition, characterized in that: include: memory and one or more processors; One or more application programs are stored in the memory, and the one or more application programs are suitable for being executed by the one or more processors to implement the smart sofa control method based on gesture recognition as described in any one of claims 1 to 5.

Images