A control method, device, equipment and medium for a rotating structure

Abstract

The present invention relates to the field of intelligent control technology, specifically to a control method, device, equipment and medium for a rotating structure, wherein the method is: obtaining the sound information and somatosensory information of users in the current venue in real time, setting an initial monitoring area; obtaining the number of users in the current venue, setting a user label for each user and associating them; analyzing the expression information included in the sound information of each user and setting a corresponding weight coefficient; dividing the initial monitoring area into multiple sub-areas and adjusting the size and position of the initial monitoring area; extracting specific instruction information from the sound information and somatosensory information, obtaining the corresponding instruction action in the specific instruction information, associating the instruction action with the corresponding user label; based on the weight coefficient and the instruction action, controlling the rotating structure to perform the corresponding action. The present application can extract specific instruction information in real time and dynamically adjust the control strategy, making the operation of the rotating structure more accurate and intelligent, thereby optimizing the overall user experience.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a control method, device, equipment and medium for a rotating structure. Background Art

[0002] With the rapid development of science and technology, intelligent control technology has been widely used in various fields, especially in robotics, automation systems, smart homes, and especially in the industrial field. Existing rotating structure control methods mostly rely on simple physical feedback or preset control logic. These methods often cannot respond quickly and accurately to environmental changes and diversified user needs.

[0003] For example, in automated production lines, equipment usually performs tasks according to fixed time intervals or programs, lacking the ability to adapt to real-time changes on site, resulting in reduced production efficiency and waste of resources. When conditions in the production environment change, such as changes in raw materials, equipment failures, or operator intervention, the existing control system is difficult to adjust in time, resulting in unstable product quality.

[0004] Existing rotating structures, for example, in automated production lines, many existing rotating structure control systems rely on simple remote controls or fixed input devices for operation, which requires users to perform cumbersome settings during operation. Moreover, in a dynamic environment, when conditions in the production environment change, such as when multiple operators intervene, it is difficult for existing control systems to adjust the rotating structure in a timely manner, resulting in unstable product quality. Secondly, although some systems have voice recognition capabilities, their recognition accuracy and response speed still need to be improved, especially since existing technologies can only operate according to preset instruction sets and cannot achieve more comprehensive intelligent and personalized control. Summary of the invention

[0005] The main purpose of the present invention is to provide a control method, device, equipment and medium for a rotating structure, aiming to overcome the defect that the existing rotating structure control cannot meet the high-precision real-time requirements.

[0006] In order to achieve the above-mentioned invention problem, the present invention proposes a control method for a rotating structure, the method comprising:

[0007] Acquire the sound information and body feeling information of users in the current venue in real time, and set an initial monitoring area based on the sound information and body feeling information;

[0008] Based on the sound information, information about the number of users in the current venue is obtained, and a user tag is set for each user, and the user tag is associated with the corresponding sound information;

[0009] Analyze the expression information included in the voice information of each user, and set a corresponding weight coefficient for each user tag based on the expression information;

[0010] Dividing the initial monitoring area into a plurality of sub-areas, each of the sub-areas corresponds to one of the user tags, and adjusting the size and position of the initial monitoring area based on the user tags;

[0011] Extracting specific instruction information from the sound information and the somatosensory information, acquiring a corresponding instruction action in the specific instruction information based on a preset instruction library, and associating the instruction action with the corresponding user tag;

[0012] Based on the weight coefficient and the command action, the rotating structure is controlled to perform a corresponding action.

[0013] Furthermore, the step of acquiring the sound information and body sensory information of users in the current venue in real time and setting an initial monitoring area based on the sound information and body sensory information includes:

[0014] The sensor on the rotating structure collects the user's voice information and body sensory information in real time, wherein the voice information includes voiceprint data, and the body sensory information includes motion trajectory data;

[0015] Based on converting the voiceprint data and motion trajectory data into voiceprint features and motion trajectory features;

[0016] A multidimensional feature space is constructed, and the voiceprint feature and the motion trajectory feature are mapped into the multidimensional feature space to form the initial monitoring area, wherein each point in the multidimensional feature space represents a voiceprint feature and a corresponding motion trajectory feature of a user.

[0017] Furthermore, the obtaining information about the number of users in the current venue based on the sound information includes:

[0018] Acquire the sound information of a preset duration, and convert the sound information of the preset duration into a sound feature matrix, wherein the sound feature matrix includes a plurality of sample points;

[0019] A neighborhood radius and a minimum number of samples are set, and all samples of each sample point in the neighborhood radius are calculated based on a preset distance parameter. If all samples of the current sample point in the neighborhood radius are greater than or equal to the minimum number of samples, the current sample point is determined to be a core sample point;

[0020] All sample points in the sound feature matrix are traversed, the number of core sample points is counted, and information on the number of users in the current venue is obtained according to the number of core sample points.

[0021] Furthermore, the step of analyzing the expression information included in the voice information of each user and setting a corresponding weight coefficient for each user tag based on the expression information includes:

[0022] Extracting the semantic content of each user's expression information, and obtaining time series data corresponding to the semantic content, wherein the time series data is the semantic content corresponding to each time point;

[0023] Scoring the semantic content at each time point based on a preset semantic database to obtain a scoring value of the semantic content at each time point;

[0024] Calculate the change rate between the newly added time point and the initial time point according to the difference in the score value between each newly added time point and the initial time point in each of the time series data;

[0025] Assigning a corresponding weight coefficient to each user tag according to the change rate, wherein the greater the change rate, the higher the corresponding weight coefficient;

[0026] The calculation expression is: , is the weight coefficient corresponding to the user tag, n is the total number of time points, and k is the index, indicating the time point currently being calculated. The sequence number ranges from 1 to n. For at time point The semantic content of For time point The semantic content of The rating value of The initial time point The semantic content score of is a very small positive number, is the weight factor used to adjust different time points Weight coefficient degree of impact.

[0027] Furthermore, the step of dividing the initial monitoring area into a plurality of sub-areas, each of which corresponds to one of the user tags, and adjusting the size and position of the initial monitoring area based on the user tags includes:

[0028] Acquire the boundary of the initial monitoring area, and establish a plurality of root support points based on the boundary of the initial monitoring area, each of the root support points being a sub-area;

[0029] Sorting the weight coefficients of each user, obtaining the priority of each user tag, and calculating the adjustment parameters of each root support point according to the priority of the user tag, wherein the adjustment parameters include a distance adjustment value and a direction adjustment value;

[0030] Combine the initial position of each root pivot with the corresponding adjustment parameter to generate new root pivot position coordinates;

[0031] Based on the new root support point position coordinates, the initial monitoring area is re-divided to form a plurality of new sub-areas, and the initial monitoring area is adjusted according to the size and position of each of the new sub-areas.

[0032] Furthermore, the step of extracting specific instruction information from the sound information and the somatosensory information, acquiring a corresponding instruction action in the specific instruction information based on a preset instruction library, and associating the instruction action with the corresponding user tag includes:

[0033] Match the voiceprint features and the corresponding motion trajectory features with the command features in the preset command library to identify specific command information;

[0034] Associating the identified specific instruction information with the user tag, dividing the priority of the instruction action according to the type of the specific instruction information, and assigning a corresponding execution weight to each instruction action;

[0035] If it is detected that an abnormality occurs during the execution of the instruction action or a cancellation instruction action occurs, the execution weight of the corresponding instruction action is adjusted based on the abnormality or the cancellation instruction action.

[0036] Further, the step of controlling the rotating structure to perform a corresponding action based on the weight coefficient and the command action includes:

[0037] Calculating the contribution of each user tag to the rotation structure action according to the weight coefficient of each user tag and the associated instruction action;

[0038] The contribution of all user tags is combined to obtain the comprehensive contribution;

[0039] A final action of the rotating structure is set according to the comprehensive contribution, and the final action at least includes a rotation speed and a rotation direction of the rotating structure.

[0040] The present invention also provides a control device for a rotating structure, comprising:

[0041] An acquisition module, used to acquire the sound information and body sensory information of users in the current venue in real time, and set an initial monitoring area based on the sound information and body sensory information;

[0042] A setting module, used for obtaining information on the number of users in the current venue based on the sound information, setting a user tag for each user, and associating the user tag with the corresponding sound information;

[0043] An analysis module, used to analyze the expression information included in the voice information of each user, and set a corresponding weight coefficient for each user tag based on the expression information;

[0044] An adjustment module, configured to divide the initial monitoring area into a plurality of sub-areas, each of which corresponds to one of the user tags, and adjust the size and position of the initial monitoring area based on the user tags;

[0045] an associating module, configured to extract specific instruction information from the sound information and the somatosensory information, obtain a corresponding instruction action from the specific instruction information based on a preset instruction library, and associate the instruction action with the corresponding user tag;

[0046] A control module is used to control the rotating structure to perform a corresponding action based on the weight coefficient and the command action.

[0047] The present invention further provides a computer device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0048] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented.

[0049] Beneficial effects:

[0050] The control method of the rotating structure proposed in this application realizes real-time and accurate identification and response to user needs by comprehensively utilizing sound information and somatosensory information, significantly improves the multi-user processing capability and flexibility of the system, dynamically monitors and adjusts the initial area to adapt to environmental changes, and sets corresponding labels and weight coefficients for each user based on sound information analysis, so as to accurately understand the user's intentions and improve response efficiency. Furthermore, by dividing the monitoring area into multiple sub-areas, interference between users is avoided, ensuring that each user's needs are met independently, and specific command information can be extracted in real time and the control strategy can be adjusted dynamically, greatly improving the accuracy and efficiency of command execution. Combined with the execution of weight coefficients and command actions, the operation of the rotating structure is more accurate and intelligent, thereby optimizing the overall user experience. Overall, the innovations in real-time, adaptability and user interaction have greatly enhanced the practicality and application potential of intelligent control technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the steps of a control method for a rotating structure in one embodiment of the present invention;

[0052] Figure 2is a schematic block diagram of a control device for a rotating structure according to an embodiment of the present invention;

[0053] Figure 3 is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention;

[0054] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "above", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of features, integers, steps, operations, elements, modules, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, components, and / or groups thereof. It should be understood that when an element is said to be "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any module and all combinations of one or more associated listed items.

[0057] Those skilled in the art will understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art in the field to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0058] Reference Figure 1 , an embodiment of the present invention provides a method for controlling a rotating structure, comprising the following steps:

[0059] S1: Acquire the sound information and body feeling information of users in the current venue in real time, and set an initial monitoring area based on the sound information and body feeling information;

[0060] In step S1, a plurality of sensors are arranged on the rotating structure, which may include at least a sound sensor and a somatosensory sensor. The sound sensor is used to capture the sound signal emitted by the user, while the somatosensory sensor is used to detect the user's body movements and postures. The sound signal includes the user's sound data, including speech content, volume and tone. The collected data needs to be preprocessed, including denoising, feature extraction, etc., to ensure the accuracy of subsequent analysis. For example, the sound data is subjected to Fourier transform to obtain its frequency components, and the different frequency components in the sound signal are identified and analyzed, so as to extract key information such as speech content, volume and tone. Preprocessing the somatosensory data includes using the data fusion technology of the accelerometer and the gyroscope to extract the effective features of the user's body movements and postures. The data fusion technology can improve the accuracy and reliability of the sensor data, so as to better understand the user's somatosensory information. Based on the sound and somatosensory data, an initial monitoring area is set. This area can be defined around the area with high user activity to ensure that most of the interactive activities can be covered. When setting the monitoring area, it can be preferred to monitor the size and position of the area based on the user's status, such as whether he is speaking or whether he is in an active state. As user behavior changes, the monitoring area also needs to be adjusted in real time. For example, when a user starts to speak or move, the monitoring area is automatically expanded or contracted to adapt to new interaction needs. In a preferred embodiment, the rotating structure includes a user interaction interface, so that the operator can monitor the current monitoring area in real time and manually adjust the area settings when necessary.

[0061] S2: obtaining information on the number of users in the current venue based on the sound information, setting a user tag for each user, and associating the user tag with the corresponding sound information;

[0062] In step S2, the sound information collected in real time is analyzed to identify the number of users in the venue, which is achieved through sound intensity, frequency and sound source localization technology, such as using a microphone array to locate the sound source and identify the existence of multiple sound sources. Each user is assigned a unique user tag based on the characteristics of the sound information (such as sound frequency, pitch, speaking style, etc.). These tags can be simple numbers (such as user 1, user 2) or more complex identifiers. Each user tag is associated with the corresponding sound information to facilitate subsequent data analysis and processing. The sound characteristics and corresponding tags of each user will be stored in the database for quick retrieval.

[0063] S3: analyzing the expression information included in the voice information of each user, and setting a corresponding weight coefficient for each user tag based on the expression information;

[0064] In step S3, the voice information of each user is deeply analyzed to identify the expression information contained therein, which includes the analysis of the emotional tendency, intonation changes, and language style of the speech content. For example, through natural language processing technology, the user's emotional state, such as happiness, anger, or neutrality, can be identified. In addition, intonation analysis can reveal the user's emotional intensity and tone changes, and language style analysis can identify the user's personal characteristics, such as whether he tends to use formal or informal vocabulary. Based on this expression information, a weight coefficient is set for each user tag to reflect its importance and activity in the current interaction. For example, a user who is speaking and is emotionally excited may be given a higher weight coefficient to ensure that his voice information receives more attention in subsequent processing.

[0065] S4: Divide the initial monitoring area into a plurality of sub-areas, each of the sub-areas corresponds to one of the user tags, and adjust the size and position of the initial monitoring area based on the user tags;

[0066] In step S4, the initial monitoring area is first divided into multiple sub-areas, each of which corresponds to a user tag. By analyzing the weight coefficient of each user tag, the size and position of each sub-area can reflect the activity and importance of the corresponding user. For example, the sub-area corresponding to a user tag with high activity and a large weight coefficient will be larger to ensure that more interaction information of the user can be captured. At the same time, according to the user's movement and sound information, the boundaries of the sub-areas are adjusted in real time to adapt to changes in user behavior.

[0067] S5: extracting specific instruction information from the sound information and the body sensory information, acquiring a corresponding instruction action in the specific instruction information based on a preset instruction library, and associating the instruction action with the corresponding user tag;

[0068] In step S5, specific instruction information is first extracted from the collected sound and somatosensory information. The instruction information may include commands, requests or other instructions issued by the user. For example, the user may control certain functions of the rotating structure, such as starting, stopping, adjusting the direction, etc., through specific statements or gestures. The preset instruction library contains a series of possible instruction actions, which correspond to specific instruction information. By matching the instruction actions in the preset instruction library, the system can recognize and execute the user's instructions. For example, when it is detected that the user says the instruction "turn left", the corresponding instruction action is found from the preset instruction library, and it is associated with the user tag that issued the instruction, thereby realizing the control of the rotating structure.

[0069] S6: Based on the weight coefficient and the command action, control the rotating structure to perform a corresponding action.

[0070] In step S6, the weight coefficient and command action information obtained in the previous step are used to accurately control the rotation structure, and the weight coefficient of each user tag and the associated command action are comprehensively considered to determine the final action of the rotation structure. For example, if the weight coefficient of a user tag is high, and the associated command action instructs the rotation structure to turn left, then the system will give priority to this command action. When executing the action, the rotation structure will adjust its rotation speed, direction and angle according to the nature of the command action and the size of the weight coefficient to ensure the smooth and accurate execution of the action. When multiple users issue different instructions at the same time, they are prioritized according to the weight coefficients corresponding to each command action, and the command action with a higher weight coefficient is executed first. If the weight coefficients are the same, a voting mechanism or a method of taking turns to execute can be adopted to ensure that each user's instruction can get a reasonable response.

[0071] In summary, by comprehensively utilizing sound information and somatosensory information, real-time and accurate recognition and response to user needs are achieved, which significantly improves the system's multi-user processing capability and flexibility, dynamically monitors and adjusts the initial area to adapt to environmental changes, and sets corresponding labels and weight coefficients for each user based on sound information analysis, so as to accurately understand user intentions and improve response efficiency. Furthermore, by dividing the monitoring area into multiple sub-areas, interference between users is avoided, ensuring that each user's needs are met independently, and specific command information can be extracted in real time and the control strategy can be adjusted dynamically, greatly improving the accuracy and efficiency of command execution. Combined with the execution of weight coefficients and command actions, the operation of the rotating structure is more accurate and intelligent, thereby optimizing the overall user experience. Overall, innovations in real-time, adaptability, and user interaction have greatly enhanced the practicality and application potential of intelligent control technology.

[0072] In an application scenario embodiment, in modern intelligent manufacturing, rotating structures are widely used in automated equipment, especially in assembly lines, logistics systems, and robot operations. The rotating structure can be a mechanical device that can rotate on multiple axes, such as a rotating robot arm, a turntable conveyor, or a multi-degree-of-freedom robot arm, specifically a typical rotating robot arm, including multiple joints and connecting rods, capable of achieving complex three-dimensional spatial motion. In the assembly line, the rotating robot arm can accurately grasp, move, and place parts to improve production efficiency and quality. In the logistics system, the turntable conveyor can include an independent rotating disc that can flexibly locate, grasp, and carry materials, thereby significantly improving production efficiency and automation. In an automated electronic product assembly line, the assembly line usually needs to move components such as circuit boards, connectors, and housings between multiple workstations. By adopting the above control method, the operator's voice information and somatosensory information can be obtained in real time, and when the operator issues instructions at the work site, the rotating structure can quickly recognize and respond. Specifically, an initial monitoring area is set according to the sound and somatosensory information on site to ensure that the operator's instructions are effectively captured. Then, the expression content in the sound information is analyzed, a user tag is assigned to each user, and a weight coefficient is set for it according to the user's expression. This personalized setting ensures that the operator's instructions can be given priority and effectively reduces the occurrence of misoperation. The initial monitoring area is divided into multiple sub-areas, each of which corresponds to a user tag, so as to achieve accurate identification and tracking of each operator. According to the data collected in real time, the rotating structure can dynamically adjust its position and movement to ensure that appropriate operations, such as assembly, inspection or handling, are performed on a specific workstation. By implementing this control method for the rotating structure, production efficiency has been greatly improved. Operators no longer need to operate through a cumbersome control interface. Only simple voice commands are needed to control the movement of the rotating structure. This direct interaction method greatly shortens the operation time and improves the flexibility of the operation. Secondly, it can better adapt to changes on site. In a complex production environment, the types and assembly order of components may change frequently. The control method can adjust the working mode of the rotating structure in time to adapt to different production tasks by analyzing the user's sound and somatosensory information in real time. This rapid response capability significantly improves the overall flexibility of the production line. In addition, personalized user labels and weight coefficient design reduce errors caused by operator differences and improve safety and reliability in the production process. By accurately analyzing the instructions of each operator, the system can better understand their needs and make more precise control when performing actions.

[0073] It is worth noting that the users referred to in this embodiment are workers operating on industrial assembly lines, but the technology is also applicable to other scenarios that require rotating structure control, such as automated warehouses, smart logistics centers, etc. In these applications, the rotating structure can be an automated forklift, rotating shelf, or sorting robot used to transport goods. By adopting control methods, these devices can quickly and accurately complete the handling, sorting, and storage of goods based on the voice and body sensory information of the workers, thereby improving logistics efficiency and reducing labor costs.

[0074] In the smart home environment, the control technology of rotating structures also has broad application prospects. For example, devices such as rotating cameras, adjustable lights, rotating dining tables or automatic curtains in the smart home system can realize intelligent control of the home environment through the analysis of sound and body sensory information. Users can adjust the status of the device through simple voice commands or gestures, such as rotating the camera to monitor different areas, or adjusting the brightness and color temperature of the light to adapt to different usage scenarios. In addition, the system can also automatically adjust the device settings according to the personal preferences and behavioral habits of family members to provide more personalized services. In an optional embodiment, in the prior art, the existing electric rotating table for a dining table is a motor-driven gear set, and the gears are connected to the turntable for rotation. The control method is that an external control panel controls the start and stop, forward and reverse rotation, speed and other functions of the motor, or a signal control signal is sent through a remote control, and the internal circuit of the turntable receives the signal to control the rotation of the motor to realize the start and stop, forward and reverse rotation and other functions of the turntable. These products greatly facilitate people's dining and improve comfort. At that time, the above-mentioned prior art had one problem: it was large in size, required high power, and had high cost; the other was that the controllability was slightly insufficient. The output gear of the above-mentioned drive reduction box was coaxial with the turntable. Making larger gears or using higher-power motors increased the size and cost; in terms of controllability, because of the problem of a control panel or remote control, it was not possible for every diner to freely operate the turntable. In the rotating structure of the present application, the cylinder (pin wheel) on the rotating table replaces the output gear of the prior art, that is, the cylinder is directly connected to the motor, and there is no need to make a larger output gear with a large number of teeth, thereby saving space and reducing costs. It has a transmission ratio of more than 3500 and a driving torque of 210mm, and can achieve the driving purpose by using a very small power motor; the built-in light sensing control module and program module replace the console or remote control of the prior art to transmit signals to control the rotating table, so that the turntable can be controlled by actions such as light dialing and stopping. (Specifically, when the power is turned on, the turntable automatically rotates forward at the set speed, stops automatically for 10 seconds after 50 seconds, then reverses for 50 seconds, and then stops for 10 seconds before entering the forward cycle. After 6 minutes, if there is no input signal, the turntable automatically enters the standby state and no longer rotates; when the turntable is externally turned over by more than 5 degrees, the turntable starts to rotate in the direction of the turn, and enters the standby state after working for 6 minutes according to the above actions; during the automatic rotation of the turntable, the turntable automatically stops for 10 seconds after the external stop button is pressed for 1 second, and then resumes the original working state. The above is the basic working state. If there are different requirements, different programs can be compiled to achieve it). A clutch mechanism is designed in the gear set to realize the meshing control of forward and reverse rotation when the turntable is turned. The present invention is designed with two types of ordinary turntables and turntables with power stations. The turntable with a power station is a turntable with a power module inserted on the basis of the ordinary model. It is very convenient to use electrical equipment such as induction cookers on this turntable, and it has the function of lighting. It can even be used to charge mobile phones when they are out of power. Unplug the power module and it becomes an ordinary turntable again.

[0075] In one embodiment, the step of acquiring the sound information and somatosensory information of users in the current venue in real time and setting an initial monitoring area based on the sound information and somatosensory information includes:

[0076] The sensor on the rotating structure collects the user's voice information and body sensory information in real time, wherein the voice information includes voiceprint data, and the body sensory information includes motion trajectory data;

[0077] Based on converting the voiceprint data and motion trajectory data into voiceprint features and motion trajectory features;

[0078] A multidimensional feature space is constructed, and the voiceprint feature and the motion trajectory feature are mapped into the multidimensional feature space to form the initial monitoring area, wherein each point in the multidimensional feature space represents a voiceprint feature and a corresponding motion trajectory feature of a user.

[0079] In the above embodiment, the sound information and somatosensory information of the user in the current venue are collected in real time by installing a variety of sensors on the rotating structure. The sound information includes voiceprint data. The user's voice features are captured by a high-sensitivity microphone. The somatosensory information is obtained by obtaining the user's motion trajectory data through motion sensors (such as accelerometers and gyroscopes). The sound and somatosensory information are obtained at the same time to ensure comprehensive monitoring of the user's status. Then, based on the collected voiceprint data and motion trajectory data, data conversion is performed to convert these raw data into voiceprint features and motion trajectory features. Machine learning algorithms such as convolutional neural networks (CNNs) can be used to extract the user's unique voiceprint pattern, and the motion trajectory features can be obtained by analyzing the user's movement pattern, speed and acceleration changes. Specifically, different algorithms and models can be used to extract voiceprint features and motion trajectory features. For example, voiceprint feature extraction can use autoencoders in deep learning to learn low-dimensional representations of voiceprint data, and motion trajectory feature extraction can use recursive neural networks to process time series data to capture the dynamic characteristics of user movements. When constructing a multidimensional feature space, dimensionality reduction techniques such as principal component analysis can be used to map high-dimensional voiceprint features and motion trajectory features into a low-dimensional space for subsequent monitoring and analysis. In this multidimensional feature space, each point represents a user's voiceprint feature and the corresponding motion trajectory feature, thus forming an initial monitoring area. In this way, the user's voice and body sensory information can be effectively integrated to provide an accurate monitoring basis for the control of rotating structures.

[0080] In one embodiment, the acquiring information about the number of users in the current venue based on the sound information includes:

[0081] Acquire the sound information of a preset duration, and convert the sound information of the preset duration into a sound feature matrix, wherein the sound feature matrix includes a plurality of sample points;

[0082] A neighborhood radius and a minimum number of samples are set, and all samples of each sample point in the neighborhood radius are calculated based on a preset distance parameter. If all samples of the current sample point in the neighborhood radius are greater than or equal to the minimum number of samples, the current sample point is determined to be a core sample point;

[0083] All sample points in the sound feature matrix are traversed, the number of core sample points is counted, and information on the number of users in the current venue is obtained according to the number of core sample points.

[0084] In the above embodiment, the sound information of the preset duration is converted into a sound feature matrix. This conversion process uses signal processing technology to extract key features in the sound signal, such as frequency, amplitude and time domain features, etc. These features are organized into a matrix form, in which each column represents a sample point and each row represents the value of a certain feature. The construction of the sound feature matrix not only helps to simplify subsequent analysis, but also can effectively convert the original sound signal into a data structure that is easy to process. This structured data form makes subsequent calculations more efficient and can reduce dependence on original data. It is worth noting that within the preset duration, the collection of sound information can be continuous or segmented. Continuous collection means continuously recording sound signals throughout the entire period of time, while segmented collection means dividing the sound signal into several shorter time periods for recording within the preset duration. Regardless of which method is used, the ultimate goal is to obtain enough sound data to ensure an accurate estimate of the number of users. After completing the construction of the sound feature matrix, a neighborhood radius and a minimum number of samples are set. The neighborhood radius refers to the range in the feature space that defines the "neighbors" of a sample point, and the minimum number of samples is the number of samples that need to exist at least in the neighborhood to determine whether a sample point is considered a core sample point. Choosing the appropriate neighborhood radius and minimum number of samples is crucial to the accuracy of the analysis. The design of this step enables the algorithm to flexibly respond to different environments and user densities. For example, in a noisy environment, a larger neighborhood radius may help aggregate more sound samples, while in a quiet environment, a smaller neighborhood may be required to avoid misjudgment. Subsequently, based on the preset distance parameter, all samples in the neighborhood radius of each sample point are calculated. By calculating the distance between the sample points in the sound feature matrix, it is identified which sample points belong to the same neighborhood. A variety of distance calculation methods can be used, such as Euclidean distance, Manhattan distance, etc. The specific choice depends on the distribution characteristics of the data and the actual application requirements. In this way, similar sound features can be effectively identified and classified into the same group. After identifying all samples in the neighborhood, determine whether the current sample point is a core sample point. If the number of all samples of the sample point in the neighborhood is greater than or equal to the minimum number of samples, the sample point is regarded as a core sample point. The advantage of this step is that the robustness and accuracy of the user number estimation can be significantly improved through cluster analysis of core sample points, avoiding erroneous judgments caused by a single noise sample. Finally, traverse all sample points in the sound feature matrix, count the number of core sample points, and obtain the user number information in the current venue based on the number of core sample points. This process ultimately maps the statistical results of the core sample points to the number of users. In this way, the actual number of users in the venue can be effectively reflected. This step not only simplifies the data processing process, but also provides more reliable user estimation results through cluster analysis.

[0085] In one embodiment, the step of analyzing the expression information included in the voice information of each user and setting a corresponding weight coefficient for each user tag based on the expression information includes:

[0086] Extracting the semantic content of each user's expression information, and obtaining time series data corresponding to the semantic content, wherein the time series data is the semantic content corresponding to each time point;

[0087] Scoring the semantic content at each time point based on a preset semantic database to obtain a scoring value of the semantic content at each time point;

[0088] Calculate the change rate between the newly added time point and the initial time point according to the difference in the score value between each newly added time point and the initial time point in each of the time series data;

[0089] Assigning a corresponding weight coefficient to each user tag according to the change rate, wherein the greater the change rate, the higher the corresponding weight coefficient;

[0090] The calculation expression is: , is the weight coefficient corresponding to the user tag, n is the total number of time points, and k is the index, indicating the time point currently being calculated. The sequence number ranges from 1 to n. For at time point The semantic content of For time point The semantic content of The rating value of The initial time point The semantic content score of is a very small positive number, is the weight factor used to adjust different time points Weight coefficient degree of impact.

[0091] In the above embodiment, this embodiment aims to elaborate on the specific steps of how to analyze the expression information contained in the voice information of each user, and set the corresponding weight coefficient for each user tag based on the expression information, and finally realize the accurate analysis of user behavior and emotion by extracting and scoring semantic content, combining time series data, and dynamically adjusting the weight coefficient of the user tag. First, it starts with extracting the semantic content of the expression information of each user. At this stage, the user's voice information is analyzed and converted into structured semantic content through natural language processing technology. For example, the user's emotional expression (such as happiness, frustration, anger, etc.) and specific topics (such as work, family, interests, etc.) can be identified. Each semantic content will correspond to a time point, forming time series data, comprehensively capturing the expression changes of the user in different time periods, thereby providing rich context information for subsequent analysis. The semantic content at each time point is scored based on a preset semantic database. The preset semantic database contains a large amount of semantic content and its corresponding scoring criteria. For example, positive emotions may be given a higher score, while negative emotions are lower, which can also be adjusted according to actual conditions. This scoring process not only ensures the objectivity and consistency of the analysis, but also helps identify the emotional intensity and expression effect of users at specific time points, making it possible to quantify the data, thus laying the foundation for the subsequent weight coefficient calculation. After obtaining the score value at each time point, the change rate is calculated based on the difference in the score value between the newly added time point and the initial time in each time series data. Specifically, the change rate is obtained by calculating the difference between the score value at each newly added time point and the score value at the initial time point, and dividing this difference by the score value at the initial time point. This process allows the system to dynamically track changes in user emotions or expressions, ensuring that the user's emotional fluctuations and focus can be reflected in a timely manner. The calculation of the change rate not only provides a quantitative basis for time series analysis, but also allows the system to identify the trend of emotional changes. Subsequently, the weight coefficient corresponding to each user tag is assigned according to the change rate. The larger the change rate, the more obvious the user's emotional change, so the corresponding weight coefficient is higher. This assignment mechanism allows the user's tag to have different weights at different time points, ensuring that the user's true emotional state and needs are better reflected. In this way, the system can more accurately capture the user's behavior pattern and adjust the importance of the user's tag in a timely manner. Furthermore, in the calculation expression, the expression , indicating a time point The change in the score of relative to the initial score, including the expression , is the rate of change, indicating the time point The degree of change in ratings.

[0092] In a specific embodiment, the specific steps and logic for calculating the change rate are as follows based on the difference in the score value between the newly added time point and the initial time in each time series data:

[0093] In a specific embodiment, the score value at the initial time point is set, and the score value is used as a reference point for comparison of the score values ​​at all subsequent time points. Then, for each newly added time point, the difference between its score value and the score value at the initial time point is calculated. Then, this difference is divided by the score value at the initial time point to obtain a ratio value, that is, the rate of change. The calculation result of the rate of change will be used to evaluate the degree of change of the user's emotion or expression over time. Further, a threshold is set to determine whether the rate of change is significant. If the rate of change exceeds this threshold, it indicates that the user's emotion or expression has changed significantly. At this time, the system will adjust the weight coefficient of the relevant user tag. The adjustment range can be determined according to the size of the rate of change. The larger the rate of change, the larger the adjustment range of the weight coefficient. At the same time, an attenuation factor is introduced to gradually reduce the influence of the rate of change at the previous time point on the current weight coefficient over time, and pay more attention to the user's recent emotion or expression changes, rather than early changes. Through the above steps, the weight coefficient of each user tag can be dynamically adjusted according to the semantic content and time series data of the user's expression information. This dynamic adjustment mechanism can more accurately capture the user's behavior pattern and emotional changes, thereby providing users with more personalized services and responses.

[0094] In one embodiment, the step of dividing the initial monitoring area into a plurality of sub-areas, each of which corresponds to one of the user tags, and adjusting the size and position of the initial monitoring area based on the user tags includes:

[0095] Acquire the boundary of the initial monitoring area, and establish a plurality of root support points based on the boundary of the initial monitoring area, each of the root support points being a sub-area;

[0096] Sorting the weight coefficients of each user, obtaining the priority of each user tag, and calculating the adjustment parameters of each root support point according to the priority of the user tag, wherein the adjustment parameters include a distance adjustment value and a direction adjustment value;

[0097] Combine the initial position of each root pivot with the corresponding adjustment parameter to generate new root pivot position coordinates;

[0098] Based on the new root support point position coordinates, the initial monitoring area is re-divided to form a plurality of new sub-areas, and the initial monitoring area is adjusted according to the size and position of each of the new sub-areas.

[0099] In the above embodiment, the boundary of the initial monitoring area is obtained. It is assumed that the initial monitoring area is set as a rectangular area, and the boundary is composed of four boundary lines, namely, top, bottom, left and right. At this time, the coordinate information of the rectangular area defines the boundary of the monitoring area. Based on the obtained boundary, multiple root fulcrums are established, each of which corresponds to a sub-area. The setting of the root fulcrum can adopt a uniform distribution method or be intelligently selected according to a specific user distribution. For example, if the number of users in the monitoring area is uneven, the user-dense area can be selected as the root fulcrum through the user distribution heat map. This flexible fulcrum setting method ensures the effectiveness of each sub-area, can cover important users, and improve the comprehensiveness of monitoring. The weight coefficient of each user is sorted to obtain the priority of each user tag. The weight coefficient is a key indicator reflecting the importance and demand intensity of the user in the monitoring area. For example, some users may have a stronger demand for services in a specific time period, and the weight coefficient may be higher. By sorting the weight coefficients, the most prioritized user tags can be identified, so that the needs of these users are given priority when adjusting the sub-areas. The advantage of this step is that it can ensure that the needs of important users are responded to in a timely manner and improve the user experience of the monitoring system. With the priority of user tags, the adjustment parameters of each root fulcrum are calculated according to these priorities. The adjustment parameters include distance adjustment value and direction adjustment value. The distance adjustment value can be obtained by analyzing the actual distance between the user and the root fulcrum, the user's activity level and other data. For example, if a user is far away from the root fulcrum and has a high weight coefficient, the root fulcrum is adjusted a certain distance to the user's location. The direction adjustment value takes into account the user's activity direction. For example, if a user often moves in a specific direction, the position of the root fulcrum can also be adjusted appropriately in that direction. This adjustment method can ensure that user needs are fully considered, thereby improving the personalization level of the service. Subsequently, the initial position of each root fulcrum is combined with the corresponding adjustment parameters to generate new root fulcrum position coordinates. The new root fulcrum position can be achieved by a simple addition operation, for example, new coordinates = initial coordinates + distance adjustment value (direction vector). This method can intuitively reflect the new position of the root fulcrum, which helps to quickly realize the redivision of the area. The advantage of this process is that the new position coordinates are quickly generated, making the entire adjustment process more efficient and adapting to rapidly changing user needs. Finally, based on the new root fulcrum position coordinates, the initial monitoring area is re-divided to form multiple new sub-areas. At this time, the boundaries of each sub-area can be determined based on the new root fulcrum position using a polygon partitioning algorithm. This algorithm can ensure that each sub-area is closely related to the corresponding root fulcrum and is as evenly distributed as possible. Through this re-division, not only the dynamic adjustment of the initial monitoring area is achieved, but also the accuracy and effectiveness of the monitoring data can be ensured. The advantage of this method is that it can respond to the changing needs of users in real time and improve the flexibility and adaptability of the monitoring system.

[0100] In a specific embodiment, the calculation expression of the above content is: ,in, is the adjusted monitoring area boundary coordinate, indicating the new monitoring area, is the boundary coordinate of the initial monitoring area, indicating the monitoring area before the user is adjusted, n is the total number of users in the monitoring area, is the weight coefficient of user ii, reflecting the importance and demand intensity of the user. The higher the value, the more urgent the user's demand. is the distance adjustment value of user i relative to its corresponding root pivot, indicating that the distance of the root pivot is adjusted according to the user's needs and position. The value can be positive or negative. A positive value indicates that the distance is closer to the user's position, while a negative value indicates that the distance is farther away. is the direction vector of user i, which represents the unit vector of the adjustment direction, where The definition is , is the coordinate of user i, It is the coordinate of the corresponding root support point. The calculation of this vector ensures the correctness of the adjustment direction.

[0101] In summary, the above steps complement each other. From obtaining boundaries to redividing regions, each step fully considers user needs and regional characteristics, ensuring the efficient operation of the monitoring system. Through this embodiment, a dynamic and intelligent monitoring area adjustment mechanism can be established, which not only improves the accuracy and real-time performance of monitoring, but also significantly improves the user experience. Ultimately, this user tag-based regional adjustment strategy will provide strong technical support for various application scenarios (such as smart cities, environmental monitoring, etc.).

[0102] In one embodiment, the step of extracting specific instruction information from the sound information and the somatosensory information, obtaining the instruction action corresponding to the specific instruction information based on a preset instruction library, and associating the instruction action with the corresponding user tag includes:

[0103] Match the voiceprint features and the corresponding motion trajectory features with the command features in the preset command library to identify specific command information;

[0104] Associating the identified specific instruction information with the user tag, dividing the priority of the instruction action according to the type of the specific instruction information, and assigning a corresponding execution weight to each instruction action;

[0105] If it is detected that an abnormality occurs during the execution of the instruction action or a cancellation instruction action occurs, the execution weight of the corresponding instruction action is adjusted based on the abnormality or the cancellation instruction action.

[0106] In the above embodiment, the extracted voiceprint features and motion trajectory features are matched with the command features in the preset command library. The preset command library contains various possible commands and their corresponding features. Through this matching, the specific command information issued by the user can be accurately identified. The advantage of this process is that the double verification of voiceprint and somatosensory information can significantly improve the accuracy of command recognition and avoid the occurrence of misoperation. Once the specific command information is identified, this information is associated with the corresponding user tag, and the priority of the command action is divided according to the type of the identified specific command information. For example, the user may give similar instructions in different scenarios, but depending on the scenario, some instructions may be more urgent or important. Each command action will be assigned a corresponding execution weight to ensure that when multiple instructions are triggered at the same time, the system can give priority to the most important instructions. The advantage of this mechanism is that it enables the system to respond flexibly in complex environments and enhances the user experience. During the execution of the command, the ability to monitor the execution status of the command, if an abnormality is detected during the execution process, such as the user gesture does not match the command, or the command fails to execute correctly, will immediately respond to these changes. At this time, the execution weight of the corresponding command action is adjusted. For example, if a command has encountered problems in multiple executions in the past, the execution weight of the command will be automatically lowered, reducing the chance of it being prioritized. The system will continuously learn and adapt to user behavior and improve the reliability of command execution.

[0107] In one embodiment, the step of controlling the rotating structure to perform a corresponding action based on the weight coefficient and the command action includes:

[0108] Calculating the contribution of each user tag to the rotation structure action according to the weight coefficient of each user tag and the associated instruction action;

[0109] The contribution of all user tags is combined to obtain the comprehensive contribution;

[0110] A final action of the rotating structure is set according to the comprehensive contribution, and the final action at least includes a rotation speed and a rotation direction of the rotating structure.

[0111] In the above embodiment, each user tag corresponds to a weight coefficient, which represents the degree of influence of the tag on the action of the rotating structure. Based on the weight coefficient of each user tag and the relevant instruction action, the contribution of each user tag to the action of the rotating structure is calculated. Specifically, the following formula can be used for calculation: Contribution = weight coefficient × instruction action intensity, and the instruction action intensity can be a numerical value related to the amplitude of the instruction action made by the user. Through this calculation, the specific impact of each user tag on the action of the rotating structure can be quantified. For example, assuming that the weight coefficient of tag A is 0.6 and the instruction action intensity is 50 degrees / second, the contribution of tag A to the rotating structure is 30 degrees / second. Repeating this process can obtain the contribution of each tag. The contribution of all user tags is combined to obtain the comprehensive contribution. The comprehensive contribution can be obtained by simple addition calculation, and the formula is: Comprehensive contribution = Contribution A + Contribution B + Contribution C + Contribution D + Contribution E. The advantage of this step is that the information of all user tags is integrated, which can reflect the overall user needs and expectations, making the control of the rotating structure more intelligent and flexible. For example, if the contributions of tags A and B are 30 degrees / second and 10 degrees / second respectively, the comprehensive contribution is 40 degrees / second. Finally, the final action of the rotating structure is set according to the comprehensive contribution, and the setting of the final action not only includes the adjustment of the rotation speed, but also involves the determination of the rotation direction. To achieve this, a decision tree or a similar algorithm can be used to judge the final action of the rotating structure according to the size of the comprehensive contribution and the relative weight of each tag. For example, if the comprehensive contribution is 40 degrees / second and tag A has the highest weight, the final action may be set to rotate at a speed of 40 degrees / second in the direction indicated by tag A. By introducing user tags and their weight coefficients, more personalized and intelligent rotating structure control can be achieved to meet the specific needs of different users. Compared with the traditional fixed control method, this dynamic adjustment based on user feedback can improve the adaptability and flexibility of the system, thereby providing a better user experience.

[0112] Reference Figure 2 , a control device for a rotating structure, comprising:

[0113] An acquisition module 100 is used to acquire the sound information and body feeling information of users in the current venue in real time, and set an initial monitoring area based on the sound information and body feeling information;

[0114] A setting module 200, configured to obtain information on the number of users in the current venue based on the sound information, set a user tag for each user, and associate the user tag with the corresponding sound information;

[0115] An analysis module 300 is used to analyze the expression information included in the voice information of each user, and to set a corresponding weight coefficient for each user tag based on the expression information;

[0116] An adjustment module 400, configured to divide the initial monitoring area into a plurality of sub-areas, each of which corresponds to one of the user tags, and adjust the size and position of the initial monitoring area based on the user tags;

[0117] An association module 500 is used to extract specific instruction information from the sound information and the body sensory information, obtain a corresponding instruction action in the specific instruction information based on a preset instruction library, and associate the instruction action with the corresponding user tag;

[0118] The control module 600 is used to control the rotating structure to perform a corresponding action based on the weight coefficient and the command action.

[0119] In this embodiment, by comprehensively utilizing sound information and somatosensory information, real-time and accurate recognition and response to user needs are achieved, the multi-user processing capability and flexibility of the system are significantly improved, the initial area is dynamically monitored and adjusted to adapt to environmental changes, and based on the analysis of sound information, corresponding labels and weight coefficients are set for each user, so as to accurately understand the user's intentions and improve the response efficiency. Furthermore, by dividing the monitoring area into multiple sub-areas, interference between users is avoided, ensuring that each user's needs are met independently, and specific instruction information can be extracted in real time and the control strategy can be adjusted dynamically, greatly improving the accuracy and efficiency of instruction execution. Combined with the execution of weight coefficients and instruction actions, the operation of the rotating structure is more accurate and intelligent, thereby optimizing the overall user experience. Overall, innovations in real-time, adaptability, and user interaction have greatly enhanced the practicality and application potential of intelligent control technology.

[0120] Based on the same idea as the method in the above-mentioned embodiment, the control device of the rotating structure provided in the present application can implement the method in the above-mentioned embodiment. For the convenience of explanation, the structural schematic diagram of the device embodiment only shows the parts related to the embodiment of the present application. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0121] Reference Figure 3 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as a control method for a rotating structure. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a control method for a rotating structure is implemented.

[0122] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, a control method for a rotating structure is implemented, including the following steps: acquiring sound information and somatosensory information of users in a current venue in real time, and setting an initial monitoring area based on the sound information and somatosensory information; acquiring information on the number of users in the current venue based on the sound information, and setting a user tag for each of the users, and associating the user tag with the corresponding sound information; analyzing the expression information included in the sound information of each user, and setting a corresponding weight coefficient for each user tag based on the expression information; dividing the initial monitoring area into a plurality of sub-areas, each of the sub-areas corresponding to a user tag, and adjusting the size and position of the initial monitoring area based on the user tag; extracting specific instruction information from the sound information and the somatosensory information, acquiring the corresponding instruction action in the specific instruction information based on a preset instruction library, and associating the instruction action with the corresponding user tag; and controlling the rotating structure to perform corresponding actions based on the weight coefficient and the instruction action.

[0123] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0124] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for controlling a rotating structure, characterized in that: The control method comprises: Acquire the sound information and body feeling information of users in the current venue in real time, and set an initial monitoring area based on the sound information and body feeling information; Based on the sound information, information about the number of users in the current venue is obtained, and a user tag is set for each user, and the user tag is associated with the corresponding sound information; Analyze the expression information included in the voice information of each user, and set a corresponding weight coefficient for each user tag based on the expression information; Dividing the initial monitoring area into a plurality of sub-areas, each of the sub-areas corresponds to one of the user tags, and adjusting the size and position of the initial monitoring area based on the user tags; Extracting specific instruction information from the sound information and the somatosensory information, acquiring a corresponding instruction action in the specific instruction information based on a preset instruction library, and associating the instruction action with the corresponding user tag; Based on the weight coefficient and the command action, controlling the rotating structure to perform a corresponding action; The step of analyzing the expression information included in the voice information of each user and setting a corresponding weight coefficient for each user tag based on the expression information includes: Extracting the semantic content of each user's expression information, and obtaining time series data corresponding to the semantic content, wherein the time series data is the semantic content corresponding to each time point; Scoring the semantic content at each time point based on a preset semantic database to obtain a scoring value of the semantic content at each time point; Calculate the change rate between the newly added time point and the initial time point according to the difference in the score value between each newly added time point and the initial time point in each of the time series data; Assigning a corresponding weight coefficient to each user tag according to the change rate, wherein the greater the change rate, the higher the corresponding weight coefficient; The calculation expression is: , is the weight coefficient corresponding to the user tag, n is the total number of time points, and k is the index, indicating the time point currently being calculated. The sequence number ranges from 1 to n. For at time point The semantic content of For time point The semantic content of The rating value of The initial time point The semantic content score of is a very small positive number, is the weight factor used to adjust different time points Weight coefficient degree of impact.

2. The method for controlling a rotating structure according to claim 1, characterized in that: The step of acquiring the sound information and body feeling information of users in the current venue in real time and setting an initial monitoring area based on the sound information and body feeling information includes: The sensor on the rotating structure collects the user's voice information and body sensory information in real time, wherein the voice information includes voiceprint data, and the body sensory information includes motion trajectory data; Based on converting the voiceprint data and motion trajectory data into voiceprint features and motion trajectory features; A multidimensional feature space is constructed, and the voiceprint feature and the motion trajectory feature are mapped into the multidimensional feature space to form the initial monitoring area, wherein each point in the multidimensional feature space represents a voiceprint feature and a corresponding motion trajectory feature of a user.

3. The method for controlling a rotating structure according to claim 1, characterized in that: The acquiring information about the number of users in the current venue based on the sound information includes: Acquire the sound information of a preset duration, and convert the sound information of the preset duration into a sound feature matrix, wherein the sound feature matrix includes a plurality of sample points; A neighborhood radius and a minimum number of samples are set, and all samples of each sample point in the neighborhood radius are calculated based on a preset distance parameter. If all samples of the current sample point in the neighborhood radius are greater than or equal to the minimum number of samples, the current sample point is determined to be a core sample point; All sample points in the sound feature matrix are traversed, the number of core sample points is counted, and information on the number of users in the current venue is obtained according to the number of core sample points.

4. The method for controlling a rotating structure according to claim 1, characterized in that: The step of dividing the initial monitoring area into a plurality of sub-areas, each of which corresponds to one of the user tags, and adjusting the size and position of the initial monitoring area based on the user tags comprises: Acquire the boundary of the initial monitoring area, and establish a plurality of root support points based on the boundary of the initial monitoring area, each of the root support points being a sub-area; Sorting the weight coefficients of each user, obtaining the priority of each user tag, and calculating the adjustment parameters of each root support point according to the priority of the user tag, wherein the adjustment parameters include a distance adjustment value and a direction adjustment value; Combine the initial position of each root pivot with the corresponding adjustment parameter to generate new root pivot position coordinates; Based on the new root support point position coordinates, the initial monitoring area is re-divided to form a plurality of new sub-areas, and the initial monitoring area is adjusted according to the size and position of each of the new sub-areas.

5. The control method of a rotating structure according to claim 2, characterized in that: The step of extracting specific instruction information from the sound information and the body sensory information, acquiring the instruction action corresponding to the specific instruction information based on a preset instruction library, and associating the instruction action with the corresponding user tag includes: Match the voiceprint features and the corresponding motion trajectory features with the command features in the preset command library to identify specific command information; Associating the identified specific instruction information with the user tag, dividing the priority of the instruction action according to the type of the specific instruction information, and assigning a corresponding execution weight to each instruction action; If it is detected that an abnormality occurs during the execution of the instruction action or a cancellation instruction action occurs, the execution weight of the corresponding instruction action is adjusted based on the abnormality or the cancellation instruction action.

6. The method for controlling a rotating structure according to claim 1, characterized in that: The step of controlling the rotating structure to perform a corresponding action based on the weight coefficient and the command action comprises: Calculating the contribution of each user tag to the rotation structure action according to the weight coefficient of each user tag and the associated instruction action; The contribution of all user tags is combined to obtain the comprehensive contribution; A final action of the rotating structure is set according to the comprehensive contribution, and the final action at least includes a rotation speed and a rotation direction of the rotating structure.

7. A control device for a rotating structure, applied to the control method according to any one of claims 1 to 6, characterized in that: include: An acquisition module, used to acquire the sound information and body sensory information of users in the current venue in real time, and set an initial monitoring area based on the sound information and body sensory information; A setting module, used for obtaining information on the number of users in the current venue based on the sound information, setting a user tag for each user, and associating the user tag with the corresponding sound information; An analysis module, used to analyze the expression information included in the voice information of each user, and set a corresponding weight coefficient for each user tag based on the expression information; An adjustment module, configured to divide the initial monitoring area into a plurality of sub-areas, each of which corresponds to one of the user tags, and adjust the size and position of the initial monitoring area based on the user tags; an associating module, configured to extract specific instruction information from the sound information and the somatosensory information, obtain a corresponding instruction action from the specific instruction information based on a preset instruction library, and associate the instruction action with the corresponding user tag; A control module is used to control the rotating structure to perform a corresponding action based on the weight coefficient and the command action.

8. A computer device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images