Same-scene cooperative working method and device of ultrasonic directional sound equipment

Abstract

The invention relates to the technical field of sound equipment data control, and discloses a same-scene cooperative work method and device for ultrasonic directional sound equipment, and the method comprises the steps: obtaining the scene setting information of each ultrasonic directional sound equipment, carrying out the cooperative work simulation, collecting multi-dimensional real-time data, and analyzing the interactive cooperative detection time; the method comprises the following steps: driving sound equipment to perform ultrasonic propagation detection according to a detection task, simulating scene live characteristics, predicting and analyzing scene interference characteristics through space-time correlation, performing environment interference correction and equipment self-correction according to the interference characteristics, optimizing working parameters, improving the cooperative working efficiency of the sound equipment, reducing environment interference, and optimizing the equipment performance. Stable work in a complex environment is ensured, wide application of the ultrasonic directional sound equipment technology is promoted, and the problem that the cooperative work effect of multiple ultrasonic sound equipment in the prior art is poor is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of audio data control, and in particular to a method and device for collaborative work of ultrasonic directional speakers in the same scenario. Background Art

[0002] With the development of ultrasonic directional speaker technology, the precise control and directional propagation of ultrasonic beams have been widely used in many fields (such as acoustic control, communication, sensing, etc.). In order to improve the performance of these ultrasonic directional speaker devices in complex collaborative work scenarios, it is necessary to ensure that they can work efficiently in cooperation, while avoiding interference and signal distortion. However, due to the variability of the environment, devices, and scenarios, a single ultrasonic directional speaker often has difficulty coping with the challenges in different working environments. Therefore, how to achieve the efficient collaborative work of multiple ultrasonic directional speakers, eliminate interference, and optimize device performance has become a technical problem to be solved urgently. In practical applications, multiple ultrasonic directional speakers often need to work in cooperation within the same scenario. Different ultrasonic directional speakers may interfere with each other due to factors such as position, working state, and environmental noise, thus affecting their working efficiency and performance. It is difficult for multiple ultrasonic directional speakers to achieve optimal collaborative work in complex scenarios. Summary of the Invention

[0003] The purpose of the present invention is to provide a method and device for collaborative work of ultrasonic directional speakers in the same scenario, aiming to solve the problem of poor collaborative work effect of multiple ultrasonic speakers in the prior art.

[0004] The present invention is implemented as follows. In the first aspect, the present invention provides a method for collaborative work of ultrasonic directional speakers in the same scenario, including: Obtain the scenario setting information of each ultrasonic directional speaker deployed in the collaborative work scenario, and simulate the collaborative work status of each ultrasonic directional speaker according to the scenario setting information of each ultrasonic directional speaker to obtain a collaborative work simulation model; Collect multi-dimensional live data of the collaborative work scenario, and analyze the timing of interactive collaborative detection of ultrasonic directional speakers for the multi-dimensional live data according to the collaborative work simulation model to obtain a detection task to be executed; Drive the corresponding ultrasonic directional speaker to perform ultrasonic directional propagation detection according to the detection task to be executed to obtain propagation detection data, and simulate the live scenario of the collaborative work scenario for the propagation detection data according to the collaborative work simulation model to obtain a live scenario feature distribution; Perform spatio-temporal correlation prediction on the multi-dimensional live data collected at each time node and the live scenario feature distribution obtained from each round of the detection task to be executed to obtain the scenario interference feature of the collaborative work scenario for each ultrasonic directional speaker; Pre - deploy the environmental interference correction parameters for each ultrasonic directional speaker according to the described scene interference characteristics, and deploy the device self - calibration parameters for each ultrasonic directional speaker through the signal drift self - detection mechanism of each ultrasonic directional speaker. Based on the pre - deployment of environmental interference correction parameters and the deployment of device self - calibration parameters, correct the working parameters of each ultrasonic directional speaker to enable it to work.

[0005] In a second aspect, the present invention provides a same - scene collaborative working device for ultrasonic directional speakers, which is used to implement the same - scene collaborative working method for ultrasonic directional speakers described in any one of the first aspects.

[0006] The present invention provides a same - scene collaborative working method for ultrasonic directional speakers, which has the following beneficial effects: The present invention obtains the scene setting information of each ultrasonic directional speaker and conducts collaborative working simulation, collects multi - dimensional live data and analyzes the interactive collaborative detection timing, drives the speakers to conduct ultrasonic propagation detection according to the detection task and simulates the scene live characteristics, predicts the scene interference characteristics through spatio - temporal correlation analysis, conducts environmental interference correction and device self - calibration according to the interference characteristics, optimizes the working parameters, improves the collaborative working efficiency of the speakers, reduces environmental interference, optimizes the device performance, ensures stable operation in complex environments, promotes the wide application of ultrasonic directional speaker technology, and solves the problem of poor collaborative working effect of multiple ultrasonic speakers in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is a schematic diagram of the steps of a same - scene collaborative working method for ultrasonic directional speakers provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0008] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0009] The implementation of the present invention will be described in detail below with reference to specific embodiments.

[0010] Refer to Figure 1 as shown, which is a preferred embodiment provided by the present invention.

[0011] In a first aspect, the present invention provides a same - scene collaborative working method for ultrasonic directional speakers, including: S1: Obtain the scene setting information of each ultrasonic directional speaker deployed in the collaborative working scene, and according to the scene setting information of each ultrasonic directional speaker, simulate the collaborative working conditions of each of the ultrasonic directional speakers to obtain a collaborative working simulation model; S2: Collect multi-dimensional live data of the collaborative work scenario, and perform timing analysis on the multi-dimensional live data for interactive collaborative detection of ultrasonic directional speakers according to the collaborative work simulation model, so as to obtain the detection tasks to be executed; S3: Drive the corresponding ultrasonic directional speakers to perform ultrasonic directional propagation detection according to the detection tasks to be executed, so as to obtain propagation detection data, and perform scene live simulation of the collaborative work scenario on the propagation detection data according to the collaborative work simulation model, so as to obtain the scene live feature distribution; S4: Perform spatio-temporal correlation prediction on the multi-dimensional live data collected at each time node and the scene live feature distribution obtained from the detection tasks to be executed in each round, so as to obtain the scene interference features of the collaborative work scenario for each ultrasonic directional speaker; S5: Pre-deploy environmental interference correction parameters for each ultrasonic directional speaker according to the scene interference features, and deploy device self-calibration parameters for each ultrasonic directional speaker through the signal drift self-check mechanism of each ultrasonic directional speaker, and correct the working parameters of each ultrasonic directional speaker based on the pre-deployment of environmental interference correction parameters and the deployment of device self-calibration parameters to perform work.

[0012] Specifically, in step S1 of the embodiment provided by the present invention, a sensor group such as a laser rangefinder and a 3D scanner is used to obtain the physical space data of the collaborative work scenario (such as room size, obstacle position, wall material reflection coefficient, etc.), generate the scene basic layout information (such as a three-dimensional coordinate grid map), locate each ultrasonic directional speaker (such as UWB positioning technology), and record its installation coordinates, orientation angle, coverage range and other scene positioning information; at the same time, collect the hardware parameters of the speaker (such as transmission frequency range, power, beam width, etc.) to form the speaker performance information.

[0013] More specifically, combine the scene basic layout information with the positioning and performance data of each speaker to generate the scene setting information of each speaker (for example: Speaker A is located at coordinates (x, y, z), covering a fan-shaped area, with a maximum power of P and a working frequency band of f1-f2). Based on the scene setting information of each speaker, create a corresponding speaker simulation unit (such as a virtual sound source model) in the digital twin platform to simulate its sound wave emission characteristics (such as beamforming direction, attenuation law).

[0014] More specifically, calculate the physical relationships between each audio simulation unit, including spatial vectors: the distance between devices, relative angles (affecting the sound wave superposition area), and temporal vectors: signal propagation delays (e.g., the sound wave transmission time Δt = d / c from audio A to audio B, where d is the distance and c is the speed of sound). Describe the spatio-temporal coupling relationship during device collaboration through vectors, and introduce the impact of the scene layout on acoustics in the model (such as the multipath effect caused by wall reflections and signal attenuation due to obstacle occlusion). Dynamically adjust the sound wave propagation parameters of each audio simulation unit (such as reflection coefficient correction) through finite element sound field simulation technology.

[0015] More specifically, simulate the sound field distribution when each audio works independently in the model, compare it with the measurement data (such as the sound pressure level distribution map) in the actual scene to verify the model accuracy. Simulate the scenario where multiple audios work simultaneously, predict the interference situation in the sound wave superposition area (such as standing waves, phase cancellation), and optimize the collaboration strategy by adjusting the spatio-temporal link vector parameters.

[0016] It can be understood that through three-dimensional space modeling and device parameter calibration, the model can accurately reflect the physical characteristics of the real scene (such as the position of obstacles, wall materials), avoiding acoustic interference caused by scene cognitive biases. For example, in a museum exhibition hall, the model can predict the reflection path of ultrasonic waves by the display case glass, avoiding the acoustic beam from accidentally touching sensitive exhibits.

[0017] More specifically, the spatio-temporal link vector enables the model to dynamically adjust the collaboration relationship between devices. For example, when the speed of sound of a certain audio drifts due to environmental temperature changes, the model can recalculate the signal delay parameters to maintain the synchronization of multiple devices.

[0018] More specifically, by simulating the sound field superposition effect when multiple devices work together, potential interference areas (such as the coincidence area of wave peaks and wave valleys) are identified in advance, and the emission angle or power distribution of the audio is adjusted. The task assignment based on the model can reduce redundant detections. For example, dynamically allocate detection tasks according to the environmental quality index, reducing the device energy consumption by 30% (compared with the fixed-period detection strategy).

[0019] Specifically, in step S2 of the embodiment provided by the present invention, data is collected through various sensors, including temperature and humidity sensors (such as SHT35), barometers (such as BMP280), air flow sensors (such as hot film anemometers), infrared human body sensors (detecting the position of listeners), cameras (identifying moving obstacles through computer vision), acoustic sensors, and microphone arrays (monitoring environmental noise, reflected sound waves).

[0020] More specifically, the data includes audio working object information, the heat map of the listener distribution, the moving trajectory of obstacles (such as changes in the crowd density), scene environment element information, the temperature and humidity gradient distribution, air pressure fluctuation data, and noise spectrum characteristics.

[0021] More specifically, for spatial data expansion and matrix construction, based on the sensor positions (such as deployed at the four corners and the center of the scene), Kriging interpolation is used to expand the discrete environmental data (such as temperature sampling points) into a continuous spatial distribution. The interpolated environmental parameters (temperature, humidity, air pressure) are encoded into an environmental factor prediction matrix according to a spatial grid (such as a resolution of 1m×1m). Each grid corresponds to an environmental factor prediction vector (such as [temperature, humidity, air pressure] = [25°C, 60%, 1013hPa]).

[0022] More specifically, for the analysis of interactive collaborative detection timing, a confidence score is calculated for the predicted value of each grid in the matrix (such as based on the consistency of adjacent sensor data. Scoring formula: Confidence = 1 - (standard deviation of predicted value / sensor measurement range)). Regions with low confidence (such as a score < 0.7) are marked as target regions that need to be detected preferentially. The environmental matrix is input into the collaborative work simulation model to simulate the change in the speed of sound under specific temperature and humidity conditions (formula: c = 331.4 + 0.6T + 0.0124H, where T is the temperature in °C and H is the humidity in %). The predicted beam offset is calculated (such as a 10°C increase in temperature results in a 2° beam offset).

[0023] More specifically, for task allocation and time weight analysis, and the calculation of the work environment quality index, the environmental interference (such as temperature fluctuation, obstacle density) in each sound coverage area is quantitatively scored (such as 0 - 100 points, the lower the score, the stronger the interference). Example formula: Quality index = 100 - (temperature fluctuation coefficient × 10 + obstacle density × 5).

[0024] More specifically, the area with a dense audience (high priority) is bound to the area with strong interference (high detection demand) to generate a sequence of tasks to be executed (such as Task 1: Beam calibration in Area A; Task 2: Obstacle avoidance in Area B). During non-task time periods (such as when there are no audiences at night), the detection trigger weight is calculated according to the change rate of the environmental matrix (such as the temperature changes by ΔT per hour) (formula: Weight = ΔT × confidence decay coefficient) to generate a collaborative detection weight feature distribution map.

[0025] More specifically, for the selection of detection timing and task construction, preset detection trigger conditions (such as the weight value > threshold θ or the confidence < 0.6), combined with time window optimization (to avoid high-load periods of equipment), to determine the trigger time (such as the low-load period at 2 am). According to the interference type in the trigger area (such as temperature drift, obstacle occlusion), the detection task parameters are defined (such as detection signal frequency, transmission power, receiver sensitivity).

[0026] It is understandable that through spatial interpolation and matrix verification, discrete sensor data is extended to a high-resolution environmental field, with the temperature and humidity prediction error ≤ ±5% (compared with the error of ±15% of the traditional mean method). Based on the dynamic task allocation of the environmental quality index, detection resources are focused on high-interference areas, and the detection task volume is reduced by 40% (compared with regular detection of the full scenario).

[0027] Specifically, in step S3 of the embodiment provided by the present invention, according to the type of detection task to be executed (such as sound speed calibration, obstacle detection), the optimal detection subject is selected from the collaborative work scenario. The first detection subject (transmitter) selects a device with a signal coverage range including the target area and a relatively low current load, and the second detection subject (receiver) is deployed at the edge of the target area or the high-incidence area of interference prediction, which is a device with a high-sensitivity receiving module.

[0028] More specifically, for the composite ultrasonic signal transmission and reception, the transmitter sends a composite ultrasonic signal, including a reference signal: a continuous wave with a fixed frequency (such as 40 kHz), which is used to measure the propagation delay and phase shift, and a modulation signal: a linearly frequency-modulated (LFM) or pseudo-random code (PRN) modulated pulse, which is used for multipath interference identification (such as detecting the reflection path through the correlation peak).

[0029] More specifically, the receiver records the following characteristics: time-domain characteristics: time difference of arrival (TDOA) of the signal, pulse width distortion; frequency-domain characteristics: signal attenuation rate (dB / m), Doppler frequency shift (used for moving obstacle detection).

[0030] More specifically, for the reverse analysis of propagation path interference, the received signal characteristics are input into the collaborative work simulation model. By inversely solving the acoustic wave propagation equation, the path interference factors are analyzed: sound speed correction, the actual sound speed is calculated according to the delay deviation, and the ambient temperature is inversely deduced (formula: T = (c_measured - 331.4) / 0.6 - 0.0207H); obstacle positioning, the distance of the obstacle is calculated through the time difference between multipath signals (such as the time difference Δt between the direct wave and the reflected wave) (formula: d = c×(Δt) / 2).

[0031] More specifically, for the dynamic correction of the environmental matrix, the analysis result is converted into an environmental element correction vector (such as [temperature correction +8°C, obstacle coordinates (x, y, z)]), and the data of the corresponding grid in the environmental element prediction matrix is updated. For the confidence weighted update, according to the reliability of the detection data (such as signal-to-noise ratio > 20 dB), a high weight (such as confidence 0.9) is given to the correction vector, and the original prediction value is preferentially covered.

[0032] More specifically, for the generation of the scene live feature distribution, the correction vector space is expanded. Based on the position of the correction vector, Gaussian Process Regression (GPR) is used to predict the changing trend of environmental parameters in adjacent areas, and matrix update is performed to generate an environmental element correction matrix covering the entire scene.

[0033] More specifically, for confidence fusion and feature extraction, low-confidence correction data with a confidence level < 0.6 (such as outliers affected by instantaneous noise) are removed, and the correction matrix and the original prediction matrix are weighted and fused to generate a scene live feature distribution map, including: an acoustic heat map: identifying areas with severe sound attenuation (such as behind obstacles), and an interference warning map: marking areas with abnormal fluctuations in temperature and humidity and the trajectories of moving obstacles.

[0034] It can be understood that through reverse analysis of actual detection data, the temperature prediction error is reduced from ±5°C to ±1°C, the obstacle positioning accuracy reaches 0.1m (the traditional ultrasonic positioning accuracy is about 0.5m), the environmental matrix correction response time is < 200ms, moving obstacles (such as pedestrians with a speed ≤ 2m / s) can be tracked in real time, the composite signal design can effectively distinguish multipath interference from real environmental changes, and the false alarm rate is reduced.

[0035] Specifically, in step S4 of the embodiment provided by the present invention, the live data (temperature, humidity, obstacle position) collected by different sensors and the scene live feature distribution (such as an acoustic heat map) generated by the detection task are aligned according to a unified time reference (error < 10ms), and the data are mapped to the same three-dimensional space grid coordinate system (such as a resolution of 1m × 1m × 0.5m) to ensure the consistency of the time and space dimensions.

[0036] More specifically, the multi-dimensional data (environmental parameters, detection results) of each node are encoded into a spatio-temporal tensor in chronological order (dimension: time × space grid × feature channels), and the feature channels include temperature, humidity, sound attenuation rate, obstacle density, etc. The sequence slices are used to divide the spatio-temporal tensor with a sliding window (such as a 1-hour window, step size 5 minutes) to construct time series samples for training and prediction.

[0037] More specifically, for multi-level spatio-temporal correlation feature extraction, the Fast Fourier Transform (FFT) is used to identify the periodic changes of environmental parameters (such as the daily temperature cycle caused by the start and stop of air conditioners), and the seasonal, trend, and residual terms of the data are separated through the STL (Seasonal-Trend Decomposition) algorithm to capture the long-term interference evolution law.

[0038] More specifically, for spatial dimension correlation modeling, the Moran's I index is used to quantify the spatial aggregation of environmental parameters (such as whether high-temperature areas are adjacent to high-humidity areas), and the Granger Causality test is used to determine whether interference in a certain area is caused by propagation from adjacent areas (such as the gradual change in sound speed caused by heat diffusion).

[0039] More specifically, for spatio-temporal joint feature fusion graph neural network (GNN) modeling, the scene space grid is constructed into a graph structure (nodes = grids, edges = adjacent relationships), the spatio-temporal graph convolutional network (ST-GCN) is used to extract spatio-temporal joint features, and the high-dimensional features are compressed into low-dimensional spatio-temporal correlation feature vectors through principal component analysis (PCA) or autoencoder (Autoencoder).

[0040] More specifically, for interference pattern generalization, K-means clustering is performed on the spatio-temporal correlation feature vectors to identify typical interference patterns (such as "morning peak of people flow - sudden increase in sound attenuation" and "afternoon sunlight - dominant shift in temperature gradient"), and physical meanings are defined for each type of interference pattern (such as pattern A = temperature gradient interference, pattern B = moving obstacle occlusion).

[0041] More specifically, for prediction model construction and inference, the Transformer time series prediction model is adopted, and the self-attention mechanism is used to capture long-range spatio-temporal dependencies. The input and output are defined as follows: the spatio-temporal correlation feature vectors of the past N time steps are input (for example, N = 12, representing data for the past 1 hour, with a time step of 5 minutes), and the scene interference features of the next M time steps are output (for example, M = 6, predicting the next 30 minutes). The prediction results are the interference intensity level (such as 0 - 10 levels) and type probability distribution (such as 70% probability of temperature interference and 30% probability of obstacle interference) for each sound coverage area.

[0042] It can be understood that in a shopping mall scenario, the change in the sound attenuation rate in crowded areas is predicted 15 minutes in advance, and the prediction error ≤ ±1dB (compared with the traditional threshold alarm error of ±3dB). Experiments show that the root mean square error (RMSE) of the prediction model in the temperature gradient interference scenario is reduced to 0.8°C (the baseline model is 2.5°C). According to the predicted interference features, the audio power allocation is adjusted in advance. For example, when it is predicted that a temperature gradient interference will occur in a certain area, the audio transmission power in that area is increased by 5% 10 minutes in advance, resulting in a 40% improvement in the stability of the sound beam. For example, in an airport waiting hall, the sound field distortion caused by the crowd gathering at the boarding gate is predicted, and the volume of the directional broadcast is dynamically increased to ensure the clarity of the notification. By spatio-temporal correlation analysis, superimposed interference factors (such as simultaneous temperature drift and obstacle reflection) are distinguished, and the accuracy rate > 90%, avoiding parameter overcorrection caused by misjudgment of a single factor.

[0043] Specifically, in step S5 of the embodiment provided by the present invention, the mapping of interference characteristics to correction requirements is classified into compensable interference according to the predicted scene interference characteristics (such as temperature gradient, obstacle density): the influence is offset by parameter adjustment (such as temperature drift → adjusting the transmission frequency), and non-compensable interference: avoidance or alarm is required (such as sudden obstacle → triggering beam steering). The requirements are quantified, and the sound speed compensation amount is: Δf = (c_nominal / c_actual - 1) × f_base, where c_actual is the actual sound speed, f_base is the reference frequency, and the beam steering angle is: θ = arctan(Δd / L), Δd is the obstacle offset, and L is the propagation distance.

[0044] More specifically, the parameter pre-deployment strategy is prioritized, and high-precision correction parameters (such as 0.1° angle fine-tuning) are deployed preferentially for audio systems covering high-priority areas (such as VIP seating areas). The parameter distribution mechanism sends the correction parameters to the target device in batches through a low-latency communication protocol (such as TSN time-sensitive network) to ensure that the coordination effectiveness time error is <10ms.

[0045] More specifically, signal drift self-check and equipment correction are carried out to continuously collect the frequency domain characteristics of the audio output signal (such as FFT spectrum), generate an ultrasonic working signal monitoring characteristic curve (record frequency stability, harmonic distortion rate), and emit a self-check pulse signal (such as a 40kHz single-frequency pulse) at every T time interval (such as every 30 minutes). The built-in microphone receives and analyzes the time delay and attenuation to generate an ultrasonic self-check characteristic curve.

[0046] More specifically, drift feature extraction and correction are performed, the frequency offset Δf and phase deviation Δφ (formula: Δφ = 2πΔfΔt, Δt is the time delay difference) between the working signal and the self-test signal are calculated, and the correction parameters are reversely solved. Phase coherence correction: adjust the DSP filter coefficients to make Δφ<5°, and sound beam divergence compensation: reverse the beam width change according to the attenuation curve and adjust the transducer drive voltage.

[0047] More specifically, in the cyclic test optimization, the initial correction parameters generated based on the theoretical model are preliminarily adjusted (e.g., the voltage adjustment amount ΔV=0.5V), and the self-test signal is re-transmitted to verify whether the frequency offset is reduced to within the threshold (e.g., Δf<10Hz). If it does not meet the threshold, the parameters are updated according to the gradient descent method (ΔV += 0.1×error), and the cycle is repeated until convergence.

[0048] More specifically, dual closed-loop parameter fusion and dynamic correction are performed, weights are assigned according to the credibility of the parameter sources (e.g., environmental correction parameter weight 0.7, self-correction parameter weight 0.3), and the Pareto optimal algorithm is used to balance parameter conflicts (e.g., environmental correction requires an increase in power, while self-correction recommends a reduction in power to extend life).

[0049] More specifically, for dynamic parameter loading and taking effect, without interrupting the operation of the audio, new parameters are loaded through the FPGA dynamic reconfiguration technology, the switching delay is <1 ms, and indicators such as sound pressure level and signal-to-noise ratio are monitored in real time. If the performance drops by >5% after correction, a rollback to the previous stable parameter set is triggered.

[0050] The present invention provides a method for collaborative operation of ultrasonic directional speakers in the same scenario, which has the following beneficial effects: The present invention obtains the scenario setting information of each ultrasonic directional speaker and conducts collaborative operation simulation, collects multi-dimensional live data and analyzes the timing of interactive collaborative detection, drives the speakers to perform ultrasonic propagation detection and simulate the characteristics of the scenario live according to the detection task, predicts the interference characteristics of the scenario through spatio-temporal correlation analysis, corrects environmental interference and performs device self-calibration according to the interference characteristics, optimizes the working parameters, improves the collaborative operation efficiency of the speakers, reduces environmental interference, optimizes the device performance, ensures stable operation in a complex environment, promotes the wide application of ultrasonic directional speaker technology, and solves the problem of poor collaborative operation effect of multiple ultrasonic speakers in the prior art.

[0051] Preferably, the steps of obtaining the scenario setting information of each ultrasonic directional speaker deployed in the collaborative operation scenario and simulating the collaborative operation status of each ultrasonic directional speaker according to the scenario setting information of each ultrasonic directional speaker to obtain a collaborative operation simulation model include: S11: Measuring the scenario building data of the collaborative operation scenario where the ultrasonic directional speakers are deployed through a measurement sensor group to obtain the scenario basic layout information of the collaborative operation scenario; S12: Collecting the specific setting information of the ultrasonic directional speakers deployed in the collaborative operation scenario through a measurement sensor group, and analyzing the relative position relationship of the collected specific setting information according to the scenario basic layout information to obtain the scenario positioning information of each ultrasonic directional speaker; S13: Obtaining the audio performance information of each ultrasonic directional speaker, and combining the audio performance information of each ultrasonic directional speaker with the scenario positioning information to obtain the scenario setting information of each ultrasonic directional speaker; S14: Performing digital simulation of the independent performance and scenario deployment of each ultrasonic directional speaker according to the scenario setting information of each ultrasonic directional speaker to obtain the audio simulation unit corresponding to each ultrasonic directional speaker and the spatio-temporal link vector between each audio simulation unit; S15: Perform digital feedback on the impact of scene layout on the sound performance of each of the sound simulation units and the spatio-temporal link vectors between the sound simulation units according to the scene basic layout information of the collaborative working scene, and adjust the sound performance parameters of each of the sound simulation units based on the digital feedback results; S16: Combine the sound simulation units with adjusted sound performance parameters through the spatio-temporal link vectors to obtain a collaborative working simulation model.

[0052] Specifically, through a sensor group deployed in the collaborative working scene, obtain the building data of the scene, including information such as spatial layout and object distribution. These data provide the basic layout information of the scene and are the basis for subsequent analysis. The data collected by the sensor group includes the position information of the ultrasonic directional speakers, obstacles in the environment, reflecting surfaces, etc., all of which will affect the sound propagation effect.

[0053] More specifically, collect specific setting information and relative position analysis. Collect the sound setting information, and obtain the specific setting information of each ultrasonic directional speaker deployed in the scene through the sensor group, including the working parameters, angles, frequencies, etc. of the speakers. Relative position relationship analysis. According to the basic layout information of the scene, analyze the relative position relationship of each ultrasonic directional speaker. Through this analysis, the positioning information of each speaker can be obtained.

[0054] More specifically, obtain the combination of sound performance information and scene positioning. Sound performance information acquisition. The performance information of each ultrasonic directional speaker, such as sound wave propagation characteristics, frequency response, maximum output power, etc., will be collected. Combine sound performance and scene positioning. Combine the performance information of each speaker and its positioning information in the scene to generate the complete scene setting information of each speaker.

[0055] More specifically, perform digital simulation of independent performance and scene deployment. According to the setting information of each speaker, perform digital simulation to simulate the independent performance of each speaker in the scene, including sound propagation path, intensity distribution, etc. The spatio-temporal link vectors of the sound simulation units. The spatio-temporal link vectors between each sound simulation unit represent the spatial position and time relationship between them. These vectors are the key parameters for describing the collaborative work of each speaker.

[0056] More specifically, through digital simulation, we can feedback the impact of different scene layouts on audio performance, such as the impact of reflective surfaces and obstacles. According to the feedback results, we adjust the audio performance parameters of each audio simulation unit to optimize the collaborative working effect of the audio. The audio simulation unit combination and the collaborative working simulation model are generated. According to the space-time link vector, the adjusted audio simulation units are combined to form a complete collaborative working model. The collaborative working simulation model will reflect the collaborative working status of each audio in the scene, including the interaction between the audio, the propagation effect, etc.

[0057] It can be understood that through precise digital simulation and adjustment, the performance of each ultrasonic directional speaker can be optimized so that they can achieve the best effect when working together. Through precise measurement and analysis of the basic layout information of the scene, the optimal speaker deployment position can be obtained to reduce performance losses caused by reflections, obstacles and other factors. The space-time link vector of the sound simulation unit can reflect the collaborative work between different speakers, optimize the synchronization and interactivity between speakers, and ensure that sound waves can accurately reach the predetermined area.

[0058] Preferably, the steps of collecting multi-dimensional live data of the collaborative work scene, and performing timing analysis of interactive collaborative detection of ultrasonic directional sound on the multi-dimensional live data according to the collaborative work simulation model to obtain the detection task to be performed include: S21: collecting data of the sound work object and scene environment elements of the collaborative work scene through a sensor group pre-deployed in the collaborative work scene, so as to obtain the sound work object information and scene environment element information in the collaborative work scene, wherein the sound work object information and the scene environment element information together constitute the multi-dimensional real-time data of the collaborative work scene; wherein the scene environment element information includes temperature element information, humidity element information and air pressure element information; S22: According to the location information of the sensor group deployed in the collaborative work scene, the scene environment element information is subjected to scene space expansion simulation to obtain the environmental element prediction data of each specific location of the collaborative work scene, and the environmental element prediction data of each specific location of the collaborative work scene is subjected to specified format conversion and overall normalization combination to obtain an environmental element prediction matrix composed of environmental element prediction vectors corresponding to each specific location of the collaborative work scene; S23: performing matrix vector interactive verification on the environmental factor prediction matrix to obtain a matrix confidence annotation set of the environmental factor prediction matrix, and performing a real-time simulation analysis of the interference of the ultrasonic directional sound by the scene on the environmental factor prediction matrix according to the matrix confidence annotation set to obtain the working environment quality index of each ultrasonic directional sound; S24: Use the working environment quality index of each ultrasonic directional speaker as a supervision condition to allocate the speaker working object information to the working tasks of each ultrasonic directional speaker according to the collaborative working simulation model, and obtain the to-be-executed task sequence of each ultrasonic directional speaker; S25: Initially arrange the task execution time for the to-be-executed task sequence of each ultrasonic directional speaker to obtain the initial task execution time distribution, and perform non-execution task time period execution interactive collaborative detection task trigger weight analysis on the initial task execution time distribution according to the working environment quality index of each ultrasonic directional speaker, so as to obtain the collaborative detection weight feature distribution of the collaborative working scenario in the future time period; S26: Select the timing according to the preset standard for the collaborative detection weight feature distribution to obtain the trigger time of the interactive collaborative detection task, and construct the requirements of the interactive collaborative detection task based on the trigger time to obtain the to-be-executed detection task.

[0059] Specifically, for the acquisition of the speaker working object information and the scenario environment element information, a sensor group deployed in the collaborative working scenario is used to collect the data of the speaker working object (such as the target sound source or receiving device) and the scenario environment elements (temperature, humidity, air pressure, etc.). These data together constitute the multi-dimensional live data of the collaborative working scenario, providing a basis for subsequent analysis.

[0060] More specifically, the environmental element information includes temperature, humidity, air pressure, etc. These factors affect the propagation characteristics and performance of the ultrasonic directional speaker. For the scenario space expansion simulation of the environmental element information, based on the position information of the sensor group, a simulation method is used to expand the environmental element information to the entire scenario space. By predicting the environmental elements at different positions, the environmental element prediction data at each position is obtained. For the construction of the environmental element prediction matrix, the obtained environmental element prediction data is subjected to format conversion and normalization processing to form the environmental element prediction matrix, which includes the environmental element prediction vector of each position.

[0061] More specifically, for the cross-validation of the environmental element prediction matrix and the live simulation analysis under scenario interference, the cross-validation of the matrix vectors is performed on the environmental element prediction matrix to ensure the consistency and credibility of the prediction data. The matrix confidence annotation set is generated according to the cross-validation results and is used to annotate the credibility of each prediction data. For the live simulation analysis under scenario interference, based on the confidence annotation set of the environmental elements, the live simulation analysis of the ultrasonic directional speaker under different scenario interference conditions is performed to evaluate the impact of the environment on the speaker performance, and the working environment quality index of each ultrasonic directional speaker is generated.

[0062] More specifically, the ultrasonic directional audio work task allocation and execution time scheduling, according to the ultrasonic directional audio work environment quality index, combined with the collaborative work simulation model, assign work tasks to each audio work object, the initial task execution time scheduling, according to the task priority and the working environment quality index of each audio, preliminarily arrange the task execution time of each audio, perform interactive collaborative detection task trigger weight analysis during the non-task execution time period, interactive collaborative detection task trigger weight analysis, in the non-task execution time period of the audio, analyze the trigger weight of the interactive collaborative detection task, weight analysis is based on factors including task priority, timing of collaborative work scenarios, changes in the audio environment, collaborative detection weight feature distribution, through weight analysis, the feature distribution of collaborative detection tasks in the future time period is obtained.

[0063] More specifically, according to preset standards (such as the urgency of the collaborative detection task, environmental conditions, sound status, etc.), an appropriate time is selected to trigger the collaborative detection task, and the trigger time of the interactive collaborative detection task is determined. Based on the timing selection, the specific trigger time of the interactive collaborative detection task is determined.

[0064] More specifically, based on the trigger time and task requirements, the specific needs of the interactive collaborative detection task are constructed, which include the task objectives, the audio equipment involved, the expected detection results, etc. The detection tasks to be executed are generated, and according to the constructed requirements, the collaborative detection tasks to be executed are obtained, and preparations are made to start the execution of the tasks.

[0065] It is understandable that the collaborative detection capability of the audio system is improved: through the design and timing selection of interactive collaborative detection tasks, the working conditions of ultrasonic directional audio can be detected and optimized in real time under multi-dimensional environmental conditions to ensure the best system performance. Through accurate prediction and simulation of environmental factors, we can fully understand the impact of different environmental factors on audio performance, and then optimize the performance of the audio system by adjusting the order of task execution and real-time feedback of environmental factors. Through quantitative evaluation of the working environment quality index, we can more reasonably allocate the work tasks of each audio system, and provide scientific scheduling arrangements for the task execution time and collaborative detection to avoid resource waste and task conflicts. Using sensor data and simulation feedback, the system can dynamically adjust the task execution plan according to environmental changes to ensure that the audio system in the collaborative work scenario can adapt to various emergencies or environmental changes.

[0066] Preferably, the steps of driving the corresponding ultrasonic directional sound to perform ultrasonic directional propagation detection according to the detection task to be executed to obtain propagation detection data, and performing a scene real-time simulation of the collaborative work scene on the propagation detection data according to the collaborative work simulation model to obtain the scene real-time feature distribution include: S31: parsing the task content of the detection task to be executed, and marking the ultrasonic directional speakers deployed in the collaborative work scenario as the first detection subject and the second detection subject of the interactive collaborative detection; S32: driving the first detection subject to send a composite directional ultrasonic detection signal toward the second detection subject at a corresponding time node according to the detection task to be performed, and instructing the second detection subject to collect characteristics of the effect of the composite directional ultrasonic detection signal to obtain propagation detection data; S33: performing reverse analysis of the propagation path environmental interference status of the composite directional ultrasonic detection signal on the propagation detection data according to the collaborative work simulation model, and verifying and modifying the environmental factor prediction matrix according to the reverse analysis result, so as to convert several environmental factor prediction vectors specified in the environmental factor prediction matrix into corresponding environmental factor correction vectors; S34: Perform correction vector expansion analysis on the environmental element prediction matrix according to the environmental element correction vector, and extract and combine effective information from the results of the correction vector expansion analysis in combination with the matrix confidence annotation set of the environmental element prediction matrix to obtain the actual feature distribution of the scene.

[0067] Specifically, according to the detection tasks to be performed, the specific requirements and objectives of the tasks are analyzed. Task content analysis can help determine which audio equipment is needed in the collaborative work scenario, which tasks are performed, and when the tasks are triggered. The ultrasonic directional audio specified in the collaborative work scenario is marked as the first detection subject and the second detection subject, respectively. These two audio devices perform different tasks respectively and conduct interactive collaborative detection.

[0068] More specifically, the ultrasonic directional speaker is driven to emit a composite ultrasonic detection signal. According to the requirements of the detection task to be performed, the first detection subject is driven to send a composite directional ultrasonic detection signal toward the second detection subject at the corresponding time node. The composite signal may include ultrasonic signals of different frequencies, amplitudes or waveforms, aiming to simulate multiple detection requirements in a real environment. The second detection subject receives the composite signal and collects characteristics of the signal propagation effect to obtain propagation detection data. These data will be used to analyze the signal propagation path, environmental interference and other conditions.

[0069] More specifically, for the reverse analysis of the propagation path environmental interference situation, according to the collaborative work simulation model, the propagation detection data is reversely analyzed to analyze the propagation path of the composite form directional ultrasonic signal and the possible environmental interference factors it may be affected by (such as temperature, humidity, obstacles, etc.). The reverse analysis helps to determine the possible deviations and interferences during the propagation process, so as to adjust the environmental impact factors. Using the results of the reverse analysis, the previously obtained environmental element prediction matrix is verified and corrected. By comparing the predicted data with the actual propagation detection data, the inaccurate parts in the prediction are corrected, thereby improving the accuracy of the environmental element prediction.

[0070] More specifically, for the correction and expansion analysis of the environmental element prediction matrix, by correcting the environmental element prediction matrix, environmental element correction vectors are generated. These correction vectors represent the accuracy improvement of each predicted environmental element at different scene positions. Applying the correction vectors to the entire environmental element prediction matrix for the expansion analysis of the correction vectors, through this expansion analysis, it can be ensured that the environmental elements of all positions and scene factors are accurately corrected.

[0071] More specifically, combined with the matrix confidence annotation set, valuable information is extracted from the results of the expansion analysis of the correction vectors. This information includes the impact of different environmental elements on ultrasonic wave propagation and its changes at different time and space positions. The extracted effective information is reasonably combined to obtain the scene actual situation feature distribution of the collaborative work scene. The scene actual situation feature distribution reflects the ultrasonic wave propagation characteristics in the real environment and helps to further optimize the ultrasonic wave detection task.

[0072] It can be understood that through the composite form of the directional ultrasonic detection signal and the reverse analysis, it is possible to comprehensively evaluate the environmental impact factors during the signal propagation process, thereby optimizing the propagation model and improving the accuracy and reliability of the ultrasonic wave directional detection task. The process of reverse analysis and the generation of environmental element correction vectors can timely correct the inaccurate parts in the environmental element prediction matrix. This correction can not only improve the accuracy of the scene simulation but also enhance the adaptability of the model in the dynamic environment. Through the reverse analysis and correction analysis of the propagation path, the scene actual situation features can be more accurately simulated, especially the interference situation of the ultrasonic wave directional propagation signal. This helps to understand the impact of different environmental conditions on the ultrasonic wave system and provides a more scientific decision-making basis. Based on the generation of the scene actual situation feature distribution, it can help to reasonably arrange the task allocation among different detection entities, ensuring that each ultrasonic wave directional sound device can interact and detect at the best time during collaborative work, thereby improving the efficiency of task execution.

[0073] Preferably, the step of performing spatio-temporal correlation prediction on the multi-dimensional live data collected at each time node and the scene live feature distribution obtained from the detection tasks to be executed in each round to obtain the scene interference features of the collaborative working scene for each ultrasonic directional sound includes: S41: Arrange the multi-dimensional live data at each time node and the detection tasks to be executed in each round in chronological order for temporal permutation and combination to obtain the scene feedback information sequence of the collaborative working scene; S42: Extract the characteristics of the change fluctuations and change forms of the feedback information in the time dimension of multiple time periods for the scene feedback information sequence to obtain the multi-level spatio-temporal correlation characteristics of the scene feedback information sequence; S43: Based on the multi-level spatio-temporal correlation characteristics, perform a generalization process on the scene live change pattern of the collaborative working scene to obtain the scene live change pattern of the collaborative working scene; S44: According to the scene live change pattern, perform a predictive analysis on the scene interference suffered by the scene live feature distribution during the next directional propagation work of the ultrasonic directional sound to obtain the scene interference characteristics of each ultrasonic directional sound.

[0074] Specifically, collect multi-dimensional live data (such as temperature, humidity, obstacle position, environmental noise, etc.) at different time nodes in the collaborative working scene. These data reflect the environmental changes during the propagation of ultrasonic signals. Collect the detection tasks to be executed in each round, including the trigger time, task requirements, form of the ultrasonic directional signal, etc. According to the time sequence, arrange the multi-dimensional live data at each time node and the detection tasks to be executed in each round for temporal permutation and combination. The sequence obtained in this way can effectively reflect the dynamic change process of the entire collaborative working scene, which is called the scene feedback information sequence.

[0075] More specifically, for feature extraction and construction of multi-level spatio-temporal correlation features, for the extraction of feedback information change fluctuation features: analyze the scene feedback information sequence in multiple time periods, and extract the change fluctuation features of the feedback information at different time scales. These fluctuation features can reveal the influence changes of environmental factors and task execution on the ultrasonic directional propagation signal in different time periods. For the extraction of feedback information change form features, further extract the form features of the feedback information change, such as the signal attenuation mode, environmental interference mode, spatio-temporal distribution change of tasks, etc. This helps to analyze the overall feedback dynamics of the collaborative working scene.

[0076] More specifically, by comprehensively analyzing the fluctuations and patterns of feedback information, multi-level spatio-temporal correlation features are constructed. These features can capture the complex relationship between ultrasonic directional signal propagation and environmental interference under different environmental conditions and provide data support for predicting scene changes.

[0077] More specifically, the generalization processing of the actual scene change pattern is based on the constructed multi-level spatio-temporal correlation features to generalize and process the actual scene change pattern of the collaborative work scene. This step will help us identify the regular changes of the scene in different time, space and task cycles, form a comprehensive understanding of the actual scene. In this process, pattern recognition techniques (such as clustering analysis, eigenvector analysis, etc.) are used to classify the change patterns of the scene and extract key factors. These patterns are helpful for understanding the interference environment faced by ultrasonic directional speakers and the environmental sensitivity of task execution.

[0078] More specifically, the predictive analysis of scene interference features is to predict the scene interference features that different ultrasonic directional speakers in the collaborative work scene may face during the next directional propagation operation according to the actual scene change pattern and combined with the spatio-temporal correlation features in the scene feedback information sequence. This includes interference characteristics such as signal attenuation, interference fluctuation, and environmental change. The interference characteristics of each ultrasonic directional speaker are analyzed to identify its performance degradation law under different environments for task optimization.

[0079] More specifically, through the aforementioned predictive analysis, the interference characteristics of each ultrasonic directional speaker in future operations are obtained. These characteristics may include the influence from the environment (such as temperature, humidity, obstacles, etc.) or interference from other audio devices, signal overlap, etc. According to the interference characteristics, an interference suppression strategy is designed for each ultrasonic directional speaker to ensure that the device can minimize the interference impact to the greatest extent and improve the execution efficiency of the detection task in the next round of tasks.

[0080] It is understandable that through the extraction of multi-level spatio-temporal correlation features, the system can more accurately understand the impact of the environment on the directional propagation of ultrasonic waves, provide accurate scene interference prediction, which helps to identify potential interference sources in advance, so as to prepare for subsequent operations. By predicting and adjusting the interference characteristics, the ultrasonic directional speaker can maintain efficient operation in a changing environment. The system will dynamically adapt to the environmental changes at different time nodes, optimize the task execution strategy, and improve the stability and accuracy of the speaker system in collaborative work. Based on spatio-temporal correlation prediction analysis, accurate information about scene changes can be provided before task execution, which can help to allocate resources reasonably, adjust task priorities, and optimize the detection path, so as to maximize the execution efficiency and accuracy of the task. By understanding the scene change pattern, the system can dynamically respond to environmental changes. For example, if the system predicts that the interference will increase in certain areas, it can adjust the transmission frequency or direction of the speaker in advance to adapt to the new interference conditions and avoid performance degradation. In a collaborative work scenario, the system can avoid signal conflicts or interference between devices by predicting the interference characteristics of each ultrasonic directional speaker, optimize the cooperation between speakers, and enable them to achieve the best effect without interference when performing common tasks.

[0081] Preferably, the steps of pre-deploying the environmental interference correction parameters for each ultrasonic directional speaker according to the scene interference characteristics and deploying the device self-correction parameters for each ultrasonic directional speaker through the signal drift self-check mechanism of each ultrasonic directional speaker include: S51: Analyze the correction requirements for the operation of the ultrasonic directional speaker based on the scene interference characteristics according to the collaborative work simulation model to obtain the environmental interference correction requirements for each ultrasonic directional speaker; S52: Analyze the implementation methods of the requirements for the environmental interference correction requirements according to the speaker performance information of each ultrasonic directional speaker, and pre-deploy the corresponding working parameters for each ultrasonic directional speaker according to the implementation methods of the requirements; S53: Perform signal drift self-check processing on each ultrasonic directional speaker through the signal drift self-check mechanism of each ultrasonic directional speaker to obtain the device signal drift characteristics of each ultrasonic directional speaker; S54: Use the device signal drift characteristics as the supervision condition to perform cyclic test adjustment on the device working parameters of each ultrasonic directional speaker until the device self-correction of each ultrasonic directional speaker is completed.

[0082] Specifically, based on the established collaborative work simulation model, analyze the current environmental conditions and interference characteristics. These environmental interference factors include temperature changes, humidity changes, changes in the position of obstacles, reflection and diffraction on the sound wave propagation path, etc. By analyzing the interference characteristics in the scenario, determine the correction requirements for each ultrasonic directional sound when it is working. For example, the correction requirements may include adjusting the transmission frequency, power, transmission angle, etc. to cope with the expected interference effects. The correction requirements for each ultrasonic directional sound will be analyzed by the model to generate a set of correction parameters for environmental interference, ensuring that the performance of each sound device during collaborative work is not affected by environmental changes.

[0083] More specifically, analyze and pre-deploy the implementation methods of environmental interference correction requirements. According to the device performance information of each ultrasonic directional sound (such as transmission frequency range, power adjustment range, working environment adaptability, etc.), analyze the implementation methods of environmental interference correction requirements. By analyzing these performance parameters, determine how the sound device specifically adjusts to meet the correction requirements. According to the requirement implementation methods, pre-deploy specific working parameters for each ultrasonic directional sound. These parameters include: selection and adjustment of the transmission frequency range, setting of the transmission power, correction of the directional angle, adaptation of the directional sound to the surrounding environment, etc., ensuring that each sound device can quickly adjust and minimize interference effects when the environment changes or the task changes, thereby improving the task execution effect.

[0084] More specifically, implement a signal drift self-check mechanism. To ensure that the sound device can maintain an accurate working state during actual operation, each ultrasonic directional sound is equipped with a signal drift self-check mechanism. This mechanism automatically detects possible signal drifts (such as frequency drift, power fluctuation, directional angle deviation, etc.) during the device's working process, and monitors the signal output changes of the sound device in real time. The sound device will perform signal drift self-check processing regularly. During the self-check process, the device will compare with known standard signals to detect whether there is a drift phenomenon. If the drift exceeds the preset allowable range, the system will trigger a self-correction process.

[0085] More specifically, through the signal drift self-check mechanism, collect and analyze the device signal drift characteristics of each ultrasonic directional sound. These characteristics include frequency drift, power change, directional error, etc., to help identify the deviation between the device's current working state and the target state. Use the device's signal drift characteristics as supervision conditions. Based on the current device deviation, adjust the working parameters of the sound device through cyclic tests. The test adjustments include adjusting the transmission power, adjusting the frequency, optimizing the directional angle, etc., until the device returns to the standard working state.

[0086] More specifically, after cyclic adjustment, when the device signal drift characteristics meet the standard working requirements, the device completes self-calibration, which ensures that each ultrasonic directional sound speaker can continuously and stably operate in a complex environment and is not affected by environmental changes or device aging.

[0087] It can be understood that by collaborating to simulate the model and analyze the correction requirements of scene interference characteristics, precise correction parameters can be provided for each ultrasonic directional sound speaker according to environmental changes. This can effectively reduce problems such as signal attenuation and offset caused by environmental interference, ensuring the stability and efficiency of the sound system. Each ultrasonic directional sound speaker can make customized adjustments to environmental interference based on its own performance information, thereby enhancing the adaptability of the device. Whether in extreme environmental conditions or when facing complex task execution scenarios, the device can quickly respond according to preset parameters, reducing unnecessary adjustment time. The introduction of the signal drift self-check mechanism enables the ultrasonic directional sound speaker to continuously monitor its signal output and automatically perform self-calibration when drift occurs. Through cyclic tests and real-time feedback adjustment, the device can automatically calibrate during long-term use, reducing manual intervention and improving the long-term stability of the device.

[0088] Preferably, the steps of performing signal drift self-check processing on each ultrasonic directional sound speaker through the signal drift self-check mechanism of each ultrasonic directional sound speaker to obtain the device signal drift characteristics of each ultrasonic directional sound speaker include: S531: Continuously monitor the audio frequency of the working signal of the ultrasonic directional sound speaker to obtain the ultrasonic working signal monitoring characteristic curve; S532: Drive the ultrasonic directional sound speaker to emit a specified form of self-check signal at predetermined time intervals and monitor the audio frequency of the self-check signal to obtain the ultrasonic self-check characteristic curve; S533: Perform curve feature fusion on the ultrasonic working signal monitoring characteristic curve and the ultrasonic self-check characteristic curve, and extract signal drift characteristics from the ultrasonic working signal monitoring characteristic curve and the ultrasonic self-check characteristic curve after curve feature fusion according to a preset signal drift evaluation standard to obtain the signal drift detection characteristics of the ultrasonic directional sound speaker at each moment; S534: Arrange the signal drift detection characteristics of the ultrasonic directional sound speaker at each moment in time sequence, and summarize the signal drift trend of the ultrasonic directional sound speaker based on the result of the time sequence arrangement to obtain the device signal drift characteristics of the ultrasonic directional sound speaker.

[0089] Specifically, during the operation of the ultrasonic directional sound device, continuously monitor the audio frequency of its working signal. By collecting and analyzing the ultrasonic signals emitted during the operation of the sound device in real time, a monitoring characteristic curve of the working signal can be obtained. This curve reflects the frequency characteristics and variation trends of the device under normal operating conditions, identifies the stability of the signal frequency, and detects whether there are frequency fluctuations or drifts caused by the device itself or environmental interference.

[0090] More specifically, for the emission and monitoring of the ultrasonic self-check signal, within a predetermined time interval, drive the ultrasonic directional sound device to emit specific self-check signals. The forms of these self-check signals may be known standard signals, such as ultrasonic signals with specific frequencies and specific waveforms. The purpose is to provide a reference standard for comparison with the working signal. Monitor the audio frequency of the self-check signals emitted by the ultrasonic directional sound device, record its frequency characteristics, and generate a self-check characteristic curve. This curve reflects the frequency performance of the device in the self-check mode and provides a stable comparison signal.

[0091] More specifically, fuse the monitoring characteristic curve of the working signal and the self-check characteristic curve. The goal of this step is to combine the frequency characteristics of the two curves, identify the frequency differences between the normal operating state and the self-check state. After fusion, it is possible to effectively distinguish the drift or deviation between the working signal and the self-check signal. Usually, a time-window or frequency-matching based method is adopted. Analyze the overlapping part and the different part of the two curves. Frequency analysis techniques, such as Fourier transform, can be used to extract their respective frequency components and conduct a comparative analysis.

[0092] More specifically, according to the preset signal drift evaluation criteria, extract the signal drift characteristics from the fused monitoring characteristic curve of the working signal and the self-check characteristic curve. These drift characteristics may include the drift amount of the frequency, the phase shift, the power fluctuation, etc. The preset criteria may include the allowable drift range (such as the maximum and minimum frequency drifts), and the trend of the drift (such as linear drift, periodic fluctuation, etc.). By comparing the actual curve with the standard, specific drift characteristics are extracted.

[0093] More specifically, the signal drift detection features at each moment are arranged in time sequence. The purpose of this step is to organize the drift features measured at different time points to form a complete signal drift time series. Based on the result of the time sequence arrangement, trend analysis is performed on the signal drift features to identify whether the drift is stable, progressive, or sudden volatility. This can help identify whether there is a gradually deteriorating signal drift or accidental interference events in the device. Through these time sequence data, the long-term trend of the device signal drift can be summarized, further providing a basis for the device status monitoring, maintenance, and repair. Finally, through the above steps, the device signal drift features are extracted and summarized. These features include: the amplitude of the signal drift (such as frequency offset, phase offset, etc.), the trend of the drift (linear, periodic, or sudden), and the degree of impact of the drift on the normal operation of the device. These signal drift features provide an important basis for subsequent device adjustment and self-correction and can help detect the health status of the device.

[0094] It can be understood that by continuously monitoring the audio frequency and comparing it with the self-check signal, the working status of the ultrasonic directional sound device can be detected in real time. This real-time detection can timely discover potential problems at the beginning of signal drift, issue early warnings, and prevent the device from having large deviations or failures. Through precise signal analysis and curve fusion, signal drift can be identified with high precision. Especially when the device is gradually aging or encountering interference, this high-precision drift detection can be adjusted before the device performance deteriorates, avoiding the unstable device performance from affecting the task execution. By extracting and summarizing the signal drift features, the system can provide specific data support for the self-correction of the device. For example, when the frequency drift exceeds the standard is detected, the system can trigger an automatic adjustment program to ensure that the device returns to the predetermined working state, thereby extending the service life of the device.

[0095] Preferably, taking the device signal drift features as the supervision condition, the step of cyclically testing and adjusting the device working parameters of each ultrasonic directional sound device until the self-correction of each ultrasonic directional sound device is completed includes: S541: Analyze the correction requirements of the ultrasonic directional sound device for the phase coherence index, beam divergence, and harmonic distortion rate according to the device signal drift features to obtain the reference correction target of the ultrasonic directional sound device; S542: Reverse-analyze and test-adjust the working correction parameters for the reference correction target to obtain the preliminary correction test effect corresponding to the reference correction target; S543: Perform repeated adjustment tests of the working correction parameters on the reference correction target according to the effect of the preliminary correction test to obtain the repeated correction test effect corresponding to the reference correction target, and perform effect feedback on the repeated adjustment test of the working correction parameters according to the repeated correction test effect to cyclically adjust the working correction parameters of the ultrasonic directional sound until the repeated correction test effect meets the reference correction target.

[0096] Specifically, by analyzing the device signal drift characteristics extracted in the foregoing steps, the correction requirements of the ultrasonic directional sound are parsed. These requirements mainly include the following aspects: detecting the phase stability of the ultrasonic signal to evaluate whether the device has phase drift or distortion; beam divergence: measuring the focusing ability and directivity of the ultrasonic beam. A larger beam divergence means that the signal propagates over a wide range and is inaccurate, affecting the directivity of the device and the sound wave transmission effect; harmonic distortion rate: detecting the purity of the ultrasonic signal. A higher harmonic distortion rate indicates that the signal generated by the device has excessive nonlinear distortion, which may affect the directivity of the ultrasonic wave and the signal quality. According to the analysis results of the signal drift characteristics, the reference correction target is determined. These targets represent the ideal state of the device, including ideal phase coherence, minimum divergence, and harmonic distortion.

[0097] More specifically, for the reverse analysis of the working correction parameters and the preliminary correction test, for a given reference correction target, it is necessary to reverse-analyze the working correction parameters. Reverse analysis is to deduce which device parameters need to be adjusted to make the ultrasonic directional sound reach the target according to the target value.

[0098] More specifically, according to the preliminary correction parameters obtained by reverse analysis, the device is preliminarily adjusted, and then an experiment is carried out to test whether the adjusted ultrasonic directional sound reaches the predetermined reference correction target, and the effect of the preliminary correction test is recorded.

[0099] More specifically, according to the effect of the preliminary correction test, analyze the gap between the correction target and the actual effect. If the correction effect does not meet the expectation, repeated adjustment tests of the working correction parameters are required. These repeated adjustments may include further optimizing the phase adjustment, improving the beam focusing, or reducing the harmonic distortion in the signal, etc. After each adjustment, a new test is carried out to test its effect.

[0100] More specifically, the results of each test are fed back to the calibration parameter adjustment strategy. This feedback mechanism can guide the further optimization of the calibration parameters, ensure the correct direction of adjustment, and gradually improve the device performance through repeated calibration tests and adjustments until the performance of the ultrasonic directional speaker fully meets the benchmark calibration target. At this time, the working parameters of the device have been finely adjusted and the self-calibration of the device is completed. This process requires continuous multi-round test adjustments to ensure that each calibration parameter (such as phase, beam divergence, and harmonic distortion) reaches the ideal state.

[0101] It can be understood that through the analysis of the calibration requirements based on the device signal drift characteristics, very precise self-calibration of the device can be achieved, and each working parameter is accurately adjusted to ensure that the device is in the best working state. Through the cyclic test adjustment and effect feedback mechanism, the system can dynamically optimize the working parameters and adjust the strategy in a timely manner according to the results of each calibration test. This dynamic optimization enables the device to adapt to different environmental changes and working conditions and continuously maintain excellent performance. After multiple rounds of adjustment and feedback, the final self-calibration result of the device can greatly improve the stability and reliability of the ultrasonic directional speaker, avoiding failures caused by signal drift or performance degradation during long-term use of the device.

[0102] In a second aspect, the present invention provides a same-scene collaborative working device for an ultrasonic directional speaker, which is used to implement the same-scene collaborative working method for an ultrasonic directional speaker according to any one of the first aspects.

[0103] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for collaborative operation of ultrasonic directional speakers in the same scenario, characterized in that, Including: Obtain the scene setting information of each ultrasonic directional speaker deployed in the collaborative work scenario, and simulate the collaborative work status of each ultrasonic directional speaker according to the scene setting information of each ultrasonic directional speaker to obtain a collaborative work simulation model; Collect multi-dimensional live data of the collaborative work scenario, and analyze the timing of interactive collaborative detection of the multi-dimensional live data by the ultrasonic directional speakers according to the collaborative work simulation model to obtain the detection tasks to be executed; Drive the corresponding ultrasonic directional speaker to perform ultrasonic directional propagation detection according to the detection task to be executed to obtain propagation detection data, and simulate the scene live condition of the collaborative work scenario according to the collaborative work simulation model for the propagation detection data to obtain the scene live condition feature distribution; Perform spatio-temporal correlation prediction on the multi-dimensional live data collected at each time node and the scene live condition feature distribution obtained from each round of the detection tasks to be executed to obtain the scene interference features of the collaborative work scenario for each ultrasonic directional speaker; Pre-deploy the environmental interference correction parameters for each ultrasonic directional speaker according to the scene interference features, and deploy the device self-correction parameters for each ultrasonic directional speaker through the signal drift self-check mechanism of each ultrasonic directional speaker. Based on the pre-deployment of environmental interference correction parameters and the deployment of device self-correction parameters, correct the working parameters of each ultrasonic directional speaker to perform the work.

2. The method for collaborative operation in the same scenario of the ultrasonic directional sound device according to claim 1, characterized in that, The steps of obtaining the scene setting information of each ultrasonic directional speaker deployed in the collaborative work scenario and simulating the collaborative work status of each ultrasonic directional speaker according to the scene setting information of each ultrasonic directional speaker to obtain a collaborative work simulation model include: Measure the scene building data of the collaborative work scenario where the ultrasonic directional speaker is deployed through a measurement sensor group to obtain the scene basic layout information of the collaborative work scenario; Collect the specific setting information of the ultrasonic directional speaker deployed in the collaborative work scenario through a measurement sensor group, and analyze the relative position relationship of the collected specific setting information according to the scene basic layout information to obtain the scene positioning information of each ultrasonic directional speaker; Obtain the acoustic performance information of each ultrasonic directional speaker, and combine the acoustic performance information of each ultrasonic directional speaker with the scene positioning information to obtain the scene setting information of each ultrasonic directional speaker; Perform digital simulation of the independent performance and scene deployment of each ultrasonic directional speaker according to the scene setting information of each ultrasonic directional speaker to obtain the acoustic simulation unit corresponding to each ultrasonic directional speaker and the spatio-temporal link vector between each acoustic simulation unit; Perform digital feedback on the influence of the scene layout on the acoustic performance for each acoustic simulation unit and the spatio-temporal link vector between each acoustic simulation unit according to the scene basic layout information of the collaborative work scenario, and adjust the acoustic performance parameters of each acoustic simulation unit based on the digital feedback result; Combine each audio simulation unit adjusted by the spatio-temporal link vector to obtain a collaborative working simulation model.

3. The method for collaborative operation in the same scenario of the ultrasonic directional sound device according to claim 1, wherein, Collect multi-dimensional live data of the collaborative working scenario, and analyze the timing of interactive collaborative detection of ultrasonic directional audio for the multi-dimensional live data according to the collaborative working simulation model, so as to obtain the steps of the detection task to be executed, including: Collect data of the audio working object and the scene environment elements in the collaborative working scenario through a sensor group pre-deployed in the collaborative working scenario, so as to obtain the audio working object information and the scene environment element information in the collaborative working scenario. The audio working object information and the scene environment element information together constitute the multi-dimensional live data of the collaborative working scenario; among them, the scene environment element information includes temperature element information, humidity element information, and air pressure element information; According to the position information of the sensor group deployed in the collaborative working scenario, perform an extended simulation of the scene space for the scene environment element information, so as to obtain the environmental element prediction data at specific positions in each place of the collaborative working scenario, perform a specified format conversion and overall normalization combination on the environmental element prediction data at specific positions in each place of the collaborative working scenario, and obtain an environmental element prediction matrix composed of environmental element prediction vectors corresponding to specific positions in each place of the collaborative working scenario; Perform an interactive verification of the matrix vector on the environmental element prediction matrix to obtain a matrix confidence annotation set of the environmental element prediction matrix, and perform a live simulation analysis of the ultrasonic directional audio affected by the scene according to the matrix confidence annotation set on the environmental element prediction matrix, so as to obtain the working environment quality index of each ultrasonic directional audio; Use the working environment quality index of each ultrasonic directional audio as a supervision condition, and allocate the working tasks pointing to each ultrasonic directional audio to the audio working object information according to the collaborative working simulation model, so as to obtain the task sequence to be executed for each ultrasonic directional audio; Make an initial arrangement of the task execution time for the task sequence to be executed for each ultrasonic directional audio to obtain an initial task execution time distribution, and perform an analysis of the trigger weight of the interactive collaborative detection task during the non-execution task time period on the initial task execution time distribution according to the working environment quality index of each ultrasonic directional audio, so as to obtain the collaborative detection weight characteristic distribution in the future time period of the collaborative working scenario; Select the timing according to the preset standard for the collaborative detection weight characteristic distribution to obtain the trigger time of the interactive collaborative detection task, and construct the requirements of the interactive collaborative detection task based on the trigger time to obtain the detection task to be executed.

4. The method for collaborative operation of ultrasonic directional speakers in the same scenario according to claim 3, characterized in that, Drive the corresponding ultrasonic directional audio to perform ultrasonic directional propagation detection according to the detection task to be executed to obtain propagation detection data, and perform a scene live simulation of the collaborative working scenario on the propagation detection data according to the collaborative working simulation model to obtain the steps of the scene live characteristic distribution, including: Analyze the task content of the to-be-executed detection task, and mark the specified ultrasonic directional speakers deployed in the collaborative work scenario as the first detection subject and the second detection subject for interactive collaborative detection respectively; Drive the first detection subject to emit a composite-form directional ultrasonic detection signal towards the second detection subject at the corresponding time node according to the to-be-executed detection task, and let the second detection subject collect the characteristics of the effect of the composite-form directional ultrasonic detection signal to obtain propagation detection data; Perform reverse analysis on the propagation path environmental interference status of the composite-form directional ultrasonic detection signal based on the collaborative work simulation model for the propagation detection data, and verify and modify the environmental element prediction matrix according to the results of the reverse analysis, so as to convert several specified environmental element prediction vectors in the environmental element prediction matrix into corresponding environmental element correction vectors; Conduct correction vector expansion analysis on the environmental element prediction matrix based on the environmental element correction vectors, and extract and combine the effective information from the results of the correction vector expansion analysis in combination with the matrix confidence annotation set of the environmental element prediction matrix to obtain the scene actual situation feature distribution.

5. The method for collaborative operation of ultrasonic directional speakers in the same scenario according to claim 1, wherein, The steps of performing spatio-temporal correlation prediction on the multi-dimensional actual situation data collected at each time node and the scene actual situation feature distribution obtained from each round of the to-be-executed detection task to obtain the scene interference characteristics of the collaborative work scenario for each ultrasonic directional speaker include: Arrange and combine the multi-dimensional actual situation data at each time node and the to-be-executed detection tasks in each round according to the time sequence to obtain the scene feedback information sequence of the collaborative work scenario; Extract the characteristics of the feedback information change fluctuation and the feedback information change form in the time dimension of multiple time periods for the scene feedback information sequence to obtain the multi-level spatio-temporal correlation characteristics of the scene feedback information sequence; Based on the multi-level spatio-temporal correlation characteristics, perform a generalization process on the scene actual situation change pattern of the collaborative work scenario to obtain the scene actual situation change pattern of the collaborative work scenario; According to the scene actual situation change pattern, perform predictive analysis on the scene interference suffered by the ultrasonic directional speaker during the next directional propagation work for the scene actual situation feature distribution to obtain the scene interference characteristics of each ultrasonic directional speaker.

6. The method for collaborative operation of ultrasonic directional speakers in the same scenario according to claim 1, characterized in that, The steps of pre-deploying the environmental interference correction parameters for each ultrasonic directional speaker according to the scene interference characteristics and performing device self-correction parameter deployment for each ultrasonic directional speaker through the signal drift self-check mechanism of each ultrasonic directional speaker include: Analyze the correction requirements for the operation of the ultrasonic directional speaker based on the collaborative work simulation model for the scene interference characteristics to obtain the environmental interference correction requirements for each ultrasonic directional speaker; Analyze the demand implementation methods for the environmental interference correction requirements according to the speaker performance information of each ultrasonic directional speaker, and pre-deploy the corresponding working parameters for each ultrasonic directional speaker according to the demand implementation methods; Self-check the signal drift of each ultrasonic directional sound device through the signal drift self-check mechanism of each ultrasonic directional sound device to obtain the device signal drift characteristics of each ultrasonic directional sound device; Using the device signal drift characteristics as the supervision condition, perform cyclic test adjustments on the device working parameters of each ultrasonic directional sound device until the device self-calibration of each ultrasonic directional sound device is completed.

7. The method for collaborative operation in the same scenario of the ultrasonic directional sound as claimed in claim 6, wherein, The steps of self-checking the signal drift of each ultrasonic directional sound device through the signal drift self-check mechanism of each ultrasonic directional sound device to obtain the device signal drift characteristics of each ultrasonic directional sound device include: Continuously monitor the audio frequency of the working signal of the ultrasonic directional sound device to obtain the monitoring characteristic curve of the ultrasonic working signal; Drive the ultrasonic directional sound device to emit a specified form of self-check signal at a predetermined time interval, and monitor the audio frequency of the self-check signal to obtain the ultrasonic self-check characteristic curve; Fuse the curve characteristics of the ultrasonic working signal monitoring characteristic curve and the ultrasonic self-check characteristic curve, and extract the signal drift characteristics of the ultrasonic working signal monitoring characteristic curve and the ultrasonic self-check characteristic curve after curve characteristic fusion according to the preset signal drift evaluation standard to obtain the signal drift detection characteristics of the ultrasonic directional sound device at each moment; Arrange the signal drift detection characteristics of the ultrasonic directional sound device at each moment in time sequence, and summarize the signal drift trend of the ultrasonic directional sound device based on the result of the time sequence arrangement to obtain the device signal drift characteristics of the ultrasonic directional sound device.

8. The method for collaborative operation of ultrasonic directional speakers in the same scenario according to claim 6, characterized in that The steps of using the device signal drift characteristics as the supervision condition to perform cyclic test adjustments on the device working parameters of each ultrasonic directional sound device until the device self-calibration of each ultrasonic directional sound device is completed include: Analyze the correction requirements of the phase coherence index, beam divergence, and harmonic distortion rate of the ultrasonic directional sound device according to the device signal drift characteristics to obtain the reference correction target of the ultrasonic directional sound device; Perform reverse analysis and test adjustment of the working correction parameters for the reference correction target to obtain the preliminary correction test effect corresponding to the reference correction target; According to the preliminary correction test effect, perform repeated adjustment tests on the working correction parameters for the reference correction target to obtain the repeated correction test effect corresponding to the reference correction target, and provide feedback on the effect of the repeated adjustment test of the working correction parameters according to the repeated correction test effect to cycle the adjustment of the working correction parameters of the ultrasonic directional sound device until the repeated correction test effect meets the reference correction target.

9. A co - working device for ultrasonic directional speakers in the same scenario, characterized in that, A method for collaborative operation of ultrasonic directional sound devices in the same scenario for implementing any one of claims 1-8.

Images