Surrounding directional loudspeaker system and time-sharing directional control method thereof

Abstract

The invention relates to the technical field of loudspeaker manufacturing, in particular to a surround directivity loudspeaker system and a time-sharing directivity control method of the surround directivity loudspeaker system, the surround directivity loudspeaker system directs different reflecting surfaces through time-sharing driving sound and simulates a plurality of surround sound sources, and the top of a box body is provided with a 3D laser radar which is further connected with a mechanical steering mechanism. A loudspeaker unit array, a frequency division and signal processing unit, an MEMS microphone and a control unit are arranged in the box body, the control unit is connected with the 3D laser radar, the microphone and the frequency division and signal processing unit, the output end of the frequency division and signal processing unit is connected with the loudspeaker array unit, the 3D laser radar is installed at the top of the sound box, and the MEMS microphone is installed at the bottom of the sound box. The scanning module is used for scanning and analyzing to obtain an angle theta of a listener relative to a central axis, a distance of the listener relative to a sound source and a three-dimensional figure of a sound field; compared with the prior art, the system structure is more compact, the cost is lower, the application range is expanded, and the limitation of a complex directivity optimization method in the prior art is overcome.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of loudspeaker manufacturing, in particular to a surround directional loudspeaker system capable of directing sound to different reflection surfaces through time-sharing driving, simulating multiple surround sound sources, and a time-sharing directional control method thereof. BACKGROUND

[0002] With the rapid development of technology, loudspeaker systems are increasingly widely used in various fields. However, in existing surround audio-visual scenes, physical loudspeakers need to be configured in each direction to achieve the effect of sound surround, and the existing directional loudspeaker solutions still have many problems. First, the traditional loudspeaker array is difficult to accurately control the propagation direction of sound, resulting in too large or too small sound diffusion range, affecting the listening experience. Second, the directional control effect of sound of different frequencies is inconsistent, resulting in too concentrated mid-high frequency sound or too strong low frequency sound diffusion, affecting the sound quality. In addition, the directional optimization method of the loudspeaker array in the prior art is relatively complex, and the directivity is not easy to control. In order to solve these problems, the industry has been exploring new technical solutions. Some solutions try to adjust the signal to optimize the directivity and frequency response, and others strive to achieve the function of automatically adjusting the directivity according to environmental changes or listener positions. However, these solutions often have the shortcomings of complex operation, high cost, limited scope of application, etc., and are difficult to promote in practical applications. Therefore, the market urgently needs a loudspeaker system that can accurately control the directivity of sound, has uniform frequency response, has adaptive adjustment capability, and can effectively optimize the sound effect of low frequency and mid-high frequency. Such a system should be able to automatically adjust the directivity and frequency response according to different environments and listener positions, while having strong practicality and feasibility, and being suitable for various scenes.

[0003] In order to solve the problems of inaccurate directional control of loudspeaker systems, uneven frequency response, and insufficient adaptive capability, some patent documents have been proposed: CN109040908A discloses a ring screen loudspeaker array with directivity and a control method thereof. The invention solves the problem of sound expansion of the main sound channel of the non-sound-transmitting material screen by distributing multiple loudspeaker sub-arrays around the non-sound-transmitting material screen, making the non-sound-transmitting material screen become a feasible movie screen, and enabling the loudspeaker array to achieve adjustable directivity. However, there is still room for optimization in the structural design of the loudspeaker array, and it is difficult to achieve a lighter and more compact array structure.

[0004] CN106031195A proposes a speaker system and a speaker with controlled directivity, which includes a plurality of sound transducer units distributed on two or more surfaces of a main body, each sound transducer unit is controlled by a device for horizontal / gain adjustment and delay / phase adjustment of each independent sound transducer unit depending on frequency. However, the invention still needs to be improved in the physical layout and acoustic performance of the complex gain and filtering device.

[0005] CN105190743A describes a directivity adjustment device that maintains a constant direct-to-reverberant sound energy ratio based on a detected position of a listener relative to a loudspeaker array, including a distance estimator, a directivity compensator, and an array processor. The invention can include a distance estimator that detects a distance between the loudspeaker array and the listener, and based on the detected distance, the directivity compensator calculates a directivity index from adjusting a beam produced by the loudspeaker array to maintain a predefined direct-to-reverberant sound energy ratio. However, it is difficult for this invention to adjust the directivity in real time according to the position of the listener and the sound playing situation to obtain the best sound effect.

[0006] In summary, the existing technology still has the following shortcomings: 1. Inaccurate directivity control: existing loudspeaker arrays are difficult to accurately control the propagation direction of sound, resulting in too large or too small sound diffusion range, affecting the listening experience. Traditional loudspeaker arrays need to change the placement position or orientation of the system to adjust the sound wave directivity or variable directional broadcast, which is difficult and time-consuming for loudspeaker systems placed high. 2. Non-uniform frequency response: the directivity control effect of different frequencies is inconsistent, resulting in too concentrated mid-high frequency sound or too strong low frequency sound diffusion, affecting the sound quality. For example, in existing technology, in order to give phase difference to low frequency sound, the sound pipe needs to be lengthened, resulting in large loudspeaker device. 3. Insufficient adaptive ability: existing technology cannot automatically adjust the directivity according to environmental changes or listener position, and needs to be adjusted manually, which is complex to operate. Most placement methods once set, the effective listening area is also fixed, and the user must be fixed in the area to receive sound, and once the area is left, the sound effect will deteriorate until there is no effect. 4. Insufficient optimization of low frequency and mid-high frequency: the directivity control effect of low frequency and mid-high frequency sound is poor, resulting in turbid low frequency sound and harsh mid-high frequency sound, affecting the sound quality. Existing stereo widening schemes introduce frequency response tinting, low frequency and central sound image attenuation, and transient tailing, and calibration is lengthy, there is no adaptive compensation for connection / phase errors, resulting in a decline in user experience. 5. Complex directivity optimization method: the directivity optimization method of the existing loudspeaker array is complex, and the directivity is not easy to control. This complex method increases the difficulty and cost of system implementation, limiting its application in restricted environments.

[0007] More importantly, the prior art solutions, whether based on fixed array electronic beamforming (such as CN106031195A) or dynamic sound beam following based on face recognition (such as WO2020228608A1), are limited to the signal processing level. Such purely electronic solutions have inherent physical limitations: when the horizontal deflection angle of electronic beamforming is too large (usually more than 30°), the beam sidelobes will deteriorate sharply, resulting in poor sound quality and inaccurate directivity. Therefore, the prior art cannot achieve accurate and stable sound following when the listener moves horizontally in a large angle, especially in a three-dimensional space with height changes. The market urgently needs a system solution that can break through the physical limitations of electronic beamforming and achieve true full-space and large-range surround directivity, thereby achieving time-sharing surround sound directivity effects. SUMMARY

[0008] The present application aims to solve the problem that existing purely electronic beamforming technology cannot effectively deal with the loss of sound directivity and the degradation of sound quality caused by large-scale horizontal movement (> 25°) and rapid movement (> 1m / s) of the listener, as well as the problem of how to use a single speaker system to achieve stable and accurate sound directivity tracking for a large-scale and fast-moving listener, and further create a surround sound listening experience. In view of the defect in the prior art that multiple physically separated speakers are required to achieve surround sound effects, resulting in a complex system, high cost, and inconvenience in installation, the present application provides a surround directivity speaker system and a time-sharing directivity control method that can achieve full-space horizontal and vertical, fast-response, and high-precision sound tracking.

[0009] The present application achieves the following:

[0010] A surround directivity speaker system is provided, which has a rectangular box-shaped box body, characterized in that a 3D laser radar is arranged on the top of the box body, and the box body is connected with a mechanical steering mechanism. The box body is provided with a speaker unit array, a frequency division and signal processing unit, a MEMS microphone, and a control unit. The control unit is connected with the 3D laser radar, the microphone, and the frequency division and signal processing unit. The output end of the frequency division and signal processing unit is connected with the speaker array unit.

[0011] The 3D laser radar is installed on the top of the sound box, used for scanning and analyzing the angle θ of the listener relative to the central axis and the distance relative to the sound source, and the three-dimensional figure of the sound field; the mechanical steering mechanism is also connected with the control unit, arranged outside the entire box body, used for adjusting the orientation of the entire box body, suitable for the occasion where the sound pointing angle deviates from the normal angle of the box body by more than 30°, since the electronic pointing reaction is fast and the mechanical pointing reaction is slow, the upper limit of the electronic pointing capability is 30°, in order to avoid the adjusted angle suddenly exceeding 30°, triggering the occasion of mechanical adjustment, the situation of overshoot (requiring electronic pointing to exceed 30°) in a short time may occur, so there is a 5° margin, for the range exceeding 25°, the mechanical adjustment is started;

[0012] The loudspeaker unit array is composed of 8 loudspeaker units, including 2 low-frequency loudspeaker units arranged at both ends of the array, using the coil loudspeaker mode; 6 middle-high-frequency loudspeaker units arranged in the middle and in a ring shape, using the mode of coil combined with piezoelectric loudspeaker, the piezoelectric mode is suitable for occasions exceeding 10KHz, the diameter of the ring is 400±50mm, the diaphragm area of the 6 middle-high-frequency loudspeaker units is 150±20mm 2 , the sound cavity volume is 40±5cm 3 , and the sensitivity is 105±5dB;

[0013] The frequency division and signal processing unit adopts the cooperative processing architecture of digital signal processor DSP and field programmable gate array FPGA, the digital signal processor DSP is configured to perform digital frequency division processing, parametric equalization and beamforming of the audio signal, the frequency division point is obtained through the mid-position of the frequency response curve of the middle-high-frequency loudspeaker and the low-frequency loudspeaker, and the low-frequency signal below 500Hz±50Hz is delivered to the 2 low-frequency loudspeaker units, and the middle-high-frequency signal above 500Hz±50Hz is delivered to the 6 middle-high-frequency loudspeaker units; the field programmable gate array FPGA is connected with the digital signal processor DSP, forming digital pointing adjustment of the loudspeaker sound direction, receiving the parametric equalization and beamforming coefficient instructions from the DSP, the parametric equalization is used for calibrating each loudspeaker, and the beamforming is used for real-time phase and delay adjustment of the signal delivered to the 6 middle-high-frequency loudspeaker units, so as to realize the directivity adjustment of less than 25° direction;

[0014] The microphone adopts a MEMS microphone and is arranged at the center position of the six middle-high frequency loudspeaker units inside the sound box, and is used for perceiving the direct sound emitted by each loudspeaker unit in time, that is, the six loudspeakers emit sound at different times respectively to periodically sound school; the input end of the control unit is connected with the environment perception unit including the MEMS microphone and the 3D laser radar, and the output end is connected with the digital signal processor DSP in the frequency division and signal processing unit, the control unit receives and fuses the data of the environment perception unit, dynamically calculates the parameter balance and beam forming parameters required by each loudspeaker unit in the loudspeaker array based on the fused data, and sends the parameters to the digital signal processor.

[0015] The shell is also provided with 12 through holes with a diameter of 8±1mm, which are distributed on the left, right and rear sides of the shell, four on each side, for balancing the air pressure inside and outside the box and optimizing the low-frequency radiation characteristics.

[0016] The mechanical steering mechanism is provided with a sound box connecting piece and a base, the base is provided with a universal rotation unit, the upper surface of the universal rotation unit is fixedly connected with the bottom of the box through the sound box connecting piece; the universal rotation unit is in a cylindrical shape, two pitch driving shafts arranged along the same radial direction are oppositely arranged on the side surface of the outer shell of the universal rotation unit, the open ends of the U-shaped supports are respectively connected with the two pitch driving shafts, the closed ends of the U-shaped supports are connected with the output shafts of the first stepping motors, and the first stepping motors are fixed on the base through supporting vertical plates; the outer shell side of the universal rotation unit is also provided with left-right heave driving shafts, the left-right heave driving shafts are located in the radial direction perpendicular to the pitch shafts, and the left-right heave driving shafts are connected with the output shafts of the second stepping motors through V-shaped rotating supports, wherein the V-shaped rotating supports are arranged below the universal rotation unit, one end of the V-shaped rotating supports is connected with the output shafts of the second stepping motors, and the second stepping motors are fixed on the base through another supporting vertical plate, and the output shafts of the second stepping motors are perpendicular to the output shafts of the first stepping motors.

[0017] The number of loudspeakers mentioned in the application is a recommended number to achieve better results, which can be appropriately increased or decreased according to actual requirements, and the size and arrangement of the single loudspeaker unit also need to be adjusted after the increase or decrease, wherein the middle-high frequency loudspeaker unit always maintains a ring structure, which is convenient for realizing arbitrary pointing, the more the loudspeaker units are, the more complex the control is, but the better the directivity is.

[0018] The application also provides a time-sharing surround directional control method for the surround directional loudspeaker system, characterized in that the control unit is further configured to perform a time-sharing surround control method: based on the three-dimensional sound field figure obtained by the 3D laser radar, the spatial relationship between the listener and the main reflecting surface in the room is established; the audio signal is time-shared and distributed to different time slices according to the virtual sound source direction; in each time slice, the loudspeaker unit array is controlled to direct the sound wave main beam to the corresponding target reflecting surface, so as to form a surround sound field at the listener position through reflected sound;

[0019] And the control unit is configured to perform the following angle adjustment cooperative control operation: when the angle between the listener position and the normal line of the front surface of the sound box is greater than a first preset threshold, the mechanical steering mechanism is controlled to adjust the orientation of the box body, and at the same time, the frequency division and signal processing unit is controlled to perform electronic beam forming compensation on the loudspeaker unit array; when the angle between the listener position and the normal line of the front surface of the sound box is reduced to below a second preset threshold, the mechanical steering mechanism is stopped, and only the frequency division and signal processing unit is used for electronic beam forming tracking.

[0020] In the application, the control unit performs an angle threshold-based electromechanical cooperative control strategy, specifically: when the angle difference between the listener position and the normal line of the front surface of the sound box is less than or equal to 25°, the system only enables electronic directional compensation adjustment, and the FPGA is used to adjust the signal of the mid-high frequency loudspeaker unit in real time, so as to realize fast and silent accurate micro-tracking; when the angle difference is greater than 25°, the system first starts the mechanical steering mechanism to coarsely position the orientation of the box body, and at the same time, the electronic directional compensation adjustment is superimposed to compensate for the delay of mechanical movement and realize smooth transition; the target of mechanical adjustment is to reduce the angle difference to within 5°, and after that, the mechanical adjustment is stopped and the electronic system is used for accurate tracking again; for the case of rapid movement of the listener (speed > 1 m / s), the mechanical and electronic systems work at the same time, and the electronic system calculates and compensates for the part of insufficient response of the mechanical system in real time; the control unit is configured to perform a 'time-sharing surround directional control method', which specifically includes:

[0021] (a) Surround sound field mapping: according to the three-dimensional information of the listening environment obtained by the 3D laser radar, a spatial relationship model of the listener position and the main reflecting surface (such as the left wall, the right wall and the back wall) in the room is established;

[0022] (b) Time-sharing channel distribution: different surround channel components in the audio signal are sequentially distributed to different time slices for processing in time;

[0023] (c) Dynamic pointing cooperative control: in each time slice, according to the target direction of the virtual sound source corresponding to the surround sound channel currently processed (such as the left surround sound channel corresponding to the specific reflection point of the left wall), the dynamic cooperative control method based on the angle threshold is called, that is, if the deviation angle θ of the target direction and the current normal direction of the system satisfies θ≤25°, the electronic beam forming is started to perform accurate pointing; if θ>25°, the mechanical steering mechanism is started to perform coarse positioning, and the electronic beam forming is superimposed to perform real-time compensation, so as to ensure that the sound wave beam is accurately pointed to the target reflection surface in the time slice;

[0024] (d) Cycle execution: in a very short time period, different surround sound channels and corresponding pointing targets are switched cyclically, and the auditory persistence effect of the human ear is used to make the listener perceive a coherent surround sound field.

[0025] The time-sharing pointing control method of the surround directional loudspeaker system in the application specifically comprises the following steps:

[0026] Step 1: initialize system configuration and set system working parameters;

[0027] Step 2: environment sensing and environment data collection: in the idle time of the loudspeaker, the sound intensity and phase difference indexes of each loudspeaker are sensed by the MEMS microphone in time sharing, the angle of the listener relative to the normal of the front face of the sound box and the distance relative to the sound source, and the 3D stereogram of the sound field are detected by the 3D laser radar in real time;

[0028] Step 3: parameter balance calculation: the digital signal processor (DSP) uses a high-precision parameter balancer to calibrate the data collected by the microphone in step 2 in the full frequency band, that is, 20Hz-20KHz, first measures the frequency response curve by using the microphone, and the deviation exceeding ±2dB is raised or attenuated to compensate for the deviation of the loudspeaker itself, solve the problem of uneven frequency response caused by standing wave and resonance, compensate for the sound intensity difference caused by the authenticity or time decay of different loudspeakers to make the deviation less than ±1dB, and calibrate the phase of different loudspeakers to make the deviation less than 0.05ms;

[0029] Step 4: According to the signal source of different sound channels, the sound direction needs to be adjusted, such as straight ahead, left, right, left rear, right rear, etc. According to the 3D stereo figure of the listener, wall or reflecting surface obtained by the 3D laser radar, the path of the sound from the sound box (center) through the wall or reflecting surface to the listener is calculated. If the change angle of the sound direction relative to the normal line of the sound box front surface needs to be adjusted within 25°, electronic pointing compensation adjustment is performed: beamforming calculation is performed, beamforming parameters are calculated by DSP according to environmental data: according to the listener position parameters including angle and distance, the time difference required for sound from each speaker unit to reach the listener is calculated, the electrical signal of each speaker unit is compensated by delay circuit or digital delay algorithm, the phase difference of each speaker unit is calculated according to the listener position and environmental parameters, based on these time difference and phase difference, the required beamforming parameters of each speaker unit are calculated, so that the sound waves at the listener position are superimposed and enhanced after being reflected by different speakers through the target reflecting point, thereby realizing the directivity of sound; if it is greater than 25°, step 6 is performed;

[0030] Step 5: Directionality fine tuning of the speaker: according to the beamforming parameters, the speaker array is controlled, the audio signal is divided into low frequency signal and medium-high frequency signal by the frequency divider, the low frequency signal is transmitted to the low frequency array composed of 2 speaker units, and the medium-high frequency signal is transmitted to the medium-high frequency array composed of 6 speaker units, the real-time delay adjustment and phase adjustment of the medium-high frequency signal are realized by FPGA high-speed parallel processing, the delay adjustment range is 0.1ms~2ms, and the phase adjustment range is -220°~220°.

[0031] Step 6: If the sound direction angle needs to be adjusted more than 25°, mechanical adjustment is first enabled, and then electronic pointing compensation adjustment is superimposed, wherein the mechanical adjustment is realized by calculating the rotation amount of the two step motors, and then driving the step motor by the control unit to drive the angle adjustment mechanism to realize the mechanical direction adjustment of the box, so as to realize large range adjustment, which can at least meet the requirement of ±60° or more.

[0032] Step 7: Return to step 2 and perform the next round of circulation.

[0033] In step 6 of the present application, the target of mechanical adjustment is within 5°, in the process of mechanical movement, the binocular camera collects the angle of the listener in real time, and the control part judges whether the sound direction angle needs to be adjusted within 5° at the same time, if not, the mechanical part continues to rotate and compensate, the electronic pointing control continues to compensate, and so on, until the angle is less than 5°, the mechanical compensation stops, and the angle compensated by the electronic pointing compensation adjustment at this time = the angle of the normal line of the sound box plane before the mechanical compensation is started - the angle adjusted by the mechanical adjustment.

[0034] In step 6 of the present application, if a wide range of sound pointing angle needs to be adjusted while the listener moves quickly, the moving speed is greater than 1 m / s, and the response speed of the mechanical part cannot keep up with the quick movement of the listener, the electronic pointing needs to be superimposed on the mechanical movement, that is, during the mechanical rotation, the mechanical control part transmits the horizontal fluctuation and vertical (inclination) rotation angle to the beam forming control part in real time (the first step motor drive generates the inclination action, which is used to adapt to the height change of the listener caused by the distance change and the standing and sitting postures of the listener; the second drive shaft generates the left-right fluctuation action, which cooperates with the first drive shaft to generate the pointing in any direction of the listener); in the electronic pointing compensation adjustment, the beam forming control part calculates the difference between the angle of the mechanical change and the angle of the actual movement of the listener obtained by the binocular camera as the compensation angle of the electronic pointing, and performs fusion processing with the distance data of the listener, recalculates the beam forming parameters and adjusts the beam forming in real time, so that the sound is always accurately pointed to the listener, the mechanical adjustment stops, and the electronic pointing compensation operation continues to maintain, which can meet the occasion of quick movement of the listener and wide range adjustment of the sound pointing angle, and in the meantime of the smooth movement of the mechanical mechanism to realize the wide range coverage, the electronic system responds at high speed to process the dynamic details, so that the tracking range and response speed far beyond the single technical limit are realized without sacrificing the sound quality.

[0035] In step 4 of the present application, the following measures are adopted:

[0036] Step 4-1: DSP is used to calculate the beam forming, the beam forming parameters are calculated according to the environmental data, and the time difference required for sound to reach the listener is calculated according to the position of the listener (for example, x = 2.0 m, y = 1.4 m), wherein the speed constant c0 = 340 m / s, and it is assumed that the distance between two adjacent speaker units is d, and it is desired to point the sound main beam to the direction with an angle of θ with the array normal, then the path length difference (i.e. the wave path difference) ΔL = d*sin(θ) is obtained when the sound from the two units reaches the listener position, in order to let the two sound waves superimpose (enhance) in phase in the θ direction, the wave path difference ΔL must be compensated, and the wavelength λ of the sound wave is c / f, and the time difference corresponding to the wave path difference ΔL is Δt = ΔL / c.

[0037] Step 4-2: FPGA needs to send signals to the unit far away from the listener in advance by Δt time, and the digital delay algorithm is used to delay and compensate the electrical signals of each speaker unit;

[0038] Step 4-3: the phase of the signal of the speaker unit far away from the listener is ahead of a specific angle, that is, the phase difference, and the conversion relationship between them is: phase difference φ = (ΔL / λ)*360°, and the phase difference of each speaker unit is calculated according to the position of the listener and the environmental parameters;

[0039] Step 4-4: Based on the time difference and phase difference, the beamforming parameters required by each speaker unit are calculated to obtain the phase compensation value of each unit.

[0040] The present application realizes a breakthrough in large-range pointing capability: through the mechanical steering mechanism, the present application first effectively solves the problem of sound directivity control in the up-down direction in practical products, enabling sound to follow the height changes of the listener's standing posture, sitting posture, etc., which is difficult to achieve by a pure electronic beamforming scheme. Through the synergy of mechanics and electricity, a performance far exceeding the limits of a single technology is achieved: mechanical adjustment overcomes the physical limits of electronic beamforming at large angles, achieving a wide-angle coverage of more than ±60°; electronic fine-tuning compensates for the deficiencies of mechanical mechanisms in precision and speed. The synergy of the two can achieve uniform coverage with a sound pressure level fluctuation of less than ±2dB within a range of ±60° in horizontal and up-down directions at a distance of 4 meters, with a tracking delay of less than 100 milliseconds.

[0041] The present application realizes a breakthrough in large-range pointing capability: through the mechanical steering mechanism, the present application first effectively solves the problem of sound directivity control in the up-down direction in practical products, enabling sound to follow the height changes of the listener's standing posture, sitting posture, etc., which is difficult to achieve by a pure electronic beamforming scheme. Through the synergy of mechanics and electricity, a performance far exceeding the limits of a single technology is achieved: mechanical adjustment overcomes the physical limits of electronic beamforming at large angles, achieving a wide-angle coverage of more than ±60°; electronic fine-tuning compensates for the deficiencies of mechanical mechanisms in precision and speed. The synergy of the two can achieve uniform coverage with a sound pressure level fluctuation of less than ±2dB within a range of ±60° in horizontal and up-down directions at a distance of 4 meters, with a tracking delay of less than 100 milliseconds.

[0042] On the basis of the mechanical and electrical synergy wide-angle pointing technology, the present application realizes a major leap in function, enabling a single-point sound source system to have the ability to simulate a surround sound field. By driving the sound to point to different reflecting surfaces at different times, the acoustic characteristics of the listening environment are effectively utilized, breaking the traditional restriction of 'one sound box one sound source', and achieving the surround sound experience that previously required a complex multi-sound box system with the simplest hardware configuration, greatly reducing the system cost and installation complexity; the present application realizes personalized adaptation of surround sound effect, and the system can intelligently calculate the optimal reflection path according to the specific structure of different rooms, so that optimized surround sound effect can be obtained in different listening environments, solving the problem of repeated calibration of traditional multi-channel systems due to differences in room acoustic environment. BRIEF DESCRIPTION OF DRAWINGS

[0043] FIG. 1 is a schematic diagram of a single-point sound source system according to the present application. Figure 1 FIG. 2 is a schematic diagram of a single-point sound source system according to the present application.

[0044] FIG. 3 is a schematic diagram of a single-point sound source system according to the present application. Figure 2is the layout diagram of the through hole on the loudspeaker system shell in the present application.

[0045] attached Figure 3 is the structural diagram of the mechanical steering mechanism in the present application, wherein Figure 3 (a) is the initial state diagram, (b) is the first stepper motor drive rotating unit tilt state diagram, (c) shows the second stepper motor drive rotating unit left and right state, combined with the tilt state of figure (b), it realizes the gimbal rotation of the box in the horizontal and tilt directions, which is the key mechanical basis for realizing the full space sound pointing.

[0046] attached Figure 4 is the structural diagram of the loudspeaker system in the present application.

[0047] attached Figure 5 is the frequency response curve before and after equalization in the embodiment of the present application.

[0048] attached Figure 6 is the phase and delay calculation table (distance 4m) of the embodiment of the present application.

[0049] attached Figure 7 is the flow chart of the present application. attached Figure 8 is the state diagram of the signal source determination of different sound channels in the present application, which needs to adjust the pointing of the sound. attached Figure 9 is the state diagram of the 3D stereogram of the listener, wall or reflecting surface obtained by the 3D laser radar in the present application.

[0050] Reference signs: speaker connecting piece 1, base 2, gimbal rotating unit 3, box 4, U-shaped support 5, first stepper motor 6, supporting vertical plate 7, second stepper motor 8, rotating support 9. DETAILED DESCRIPTION

[0051] The present application will be further described below by means of the accompanying drawings and embodiments.

[0052] The present application is directed to the prior art, the inherent limitations of a single technical means when adjusting the direction of sound waves: pure mechanical adjustment is slow, noisy and prone to wear, making it difficult to achieve smooth sound field fine tuning; pure electronic beam forming is low in efficiency at low frequencies and wide angles, and is physically limited by array size, with directional control ability decreasing sharply with frequency and increasing angle. When faced with the need for wide-angle coverage, those skilled in the art tend to design larger arrays or use multiple fixed speakers, rather than thinking of modifying the mechanical structure and deeply coupling with real-time signal processing algorithms. The adaptive technology of EAW mainly focuses on adjusting the beam electronically, and does not mention coordination with mechanical steering. EAW adaptive technology is indeed at the forefront of the industry in terms of low-frequency control, but it cannot "completely solve" the problem of low-frequency directivity. It is more of a physical limit, maximizing control ability through algorithms and array design, significantly improving low-frequency coverage and suppression under ideal conditions, but in complex sites, limited arrays or insufficient debugging, energy inequality, shielding interference and other problems may still occur.

[0053] The present application proposes a surround directivity loudspeaker system, which mainly includes a loudspeaker array, a frequency division and signal processing unit, an environment perception unit and an intelligent control unit, as shown in the accompanying Figure 4 The loudspeaker array is composed of 8 loudspeaker units, of which 2 are low-frequency loudspeakers placed at both ends, and the other 6 are mid-high frequency loudspeakers in the middle. The mid-high frequency loudspeakers are arranged in a ring array, with the 6 mid-high frequency loudspeaker units evenly distributed on a circumference with a diameter of 400 mm. This arrangement is conducive to beam forming and control in the 180° horizontal direction, with a diaphragm area of 150 mm 2 , a sound cavity volume of 40 cm 3 , and a sensitivity of 105 dB;

[0054] The frequency divider uses digital frequency division (DSP implementation), and the frequency division point is obtained by the mid-point position of the frequency response curves of the mid-high frequency loudspeakers and the low-frequency loudspeakers. For example, the audio signal is divided at 500 Hz, the low-frequency signal (below 500 Hz): delivered to a low-frequency array composed of 2 loudspeaker units, which are driven by signal preprocessing (Butterworth low-pass filter, cutoff frequency 500 Hz) and power amplifier, and are responsible for powerful low-frequency playback; the mid-high frequency signal (above 500 Hz): delivered to a mid-high frequency array composed of 6 loudspeaker units; the 6 units are controlled by FPGA, and through real-time phase and delay adjustment of the signals of each unit, precise electronic beam forming is achieved to control the directivity of mid-high frequency sound waves;

[0055] The core of the signal processing unit is a cooperative processing architecture of a DSP and an FPGA, wherein the DSP is responsible for running upper-layer algorithms, such as coefficients required by frequency division, parameter equalization, calculation of beam forming (the DSP calculates initial phase and amplitude weight required by each of the six units according to a desired beam shape (such as narrower or wider) and direction, which is equivalent to setting an “initial shape and direction” of the beam), environmental perception information fusion, system control logic and the like;

[0056] In the digital signal processor (DSP), first, the time difference required for sound to reach each loudspeaker unit is calculated according to the listener position (provided by the environmental perception unit), then, the electrical signal delivered to each unit is precisely delayed and compensated by a delay circuit or a digital delay algorithm, by changing the phase difference, the main lobe direction of the signal beam of the loudspeaker unit can be deflected to realize dynamic scanning or tracking, finally, the sound waves emitted by all units are in-phase superimposed in the target direction (at the listener), the energy is the strongest, forming a “main beam”; in other directions, they are partially offset due to different phases, and the energy is weaker, the FPGA receives the coefficient instructions from the DSP, and performs precise real-time delay and phase adjustment on the six high-frequency signals at high speed and in parallel, which is the key to realizing beam forming;

[0057] The environmental perception unit is the “eyes and ears” of the system, including: a 3D laser radar: for identifying the position of the listener, once the listener moves, the system will immediately recalculate the delay parameters required for each loudspeaker unit and dynamically adjust to ensure that the main beam always “follows” the listener; this not only greatly improves the sound clarity and immersion in the listening experience, but also effectively reduces the noise interference on the surrounding environment; the microphone is a MEMS microphone and is arranged at the center position of the six high-frequency loudspeaker units inside the sound box, for time-sharing (the six loudspeakers emit sound from time) to perceive the direct sound emitted by each loudspeaker unit, for periodic time-varying sound calibration;

[0058] The intelligent control unit is the “brain” of the system, which receives all data of the environmental perception unit, and performs fusion processing based on intelligent algorithms, the module can record the optimized parameters in different environments, so that the system can better adapt to specific environments. Dynamically calculate the best beam forming parameters (i.e. delay and phase values) and frequency response equalization parameters of the loudspeaker array in the current environment.

[0059] The mixed directivity control includes electronic adjustment and mechanical adjustment, the control signal generated by the electronic adjustment is mainly sent to the signal processing unit (FPGA), and the electronic beam steering is realized by changing the phase and delay of the electric signal. The mechanical adjustment control unit is also connected with a stepping motor driving system. The loudspeaker box is connected with a micro stepping motor, which is used to fine-tune the physical pitch and horizontal orientation angle according to the instruction of the control signal. This mechanical adjustment is mainly used to assist the electronic beam forming, compensate for the performance decline of the electronic method at extreme angles, or realize the adjustment of a larger range of coverage, which is an effective supplement to the electronic adjustment;

[0060] As shown in the accompanying drawings Figure 3 , the mechanical steering mechanism is provided with a loudspeaker connecting piece 1 and a base 2, the base 2 is provided with a universal rotation unit 3, the upper surface of the universal rotation unit 3 is fixedly connected with the bottom of the box body 4 through the loudspeaker connecting piece 1; the universal rotation unit 3 is in a cylindrical shape, two pitch drive shafts arranged along the same radial direction are oppositely arranged on the outer shell side of the universal rotation unit 3, the open ends of the U-shaped support 5 are respectively connected with the two pitch drive shafts, the closed end of the U-shaped support 5 is connected with the output shaft of the first stepping motor 6, and the first stepping motor 6 is fixed on the base 2 through a support vertical plate 7; the outer shell side of the universal rotation unit 3 is also provided with left and right heave drive shafts, the left and right heave drive shafts are located on the radial direction perpendicular to the pitch shaft, and the left and right heave drive shafts are connected with the output shaft of the second stepping motor 8 through a V-shaped rotating support 9, wherein the V-shaped rotating support 9 is located directly below the universal rotation unit 3, one end of the V-shaped rotating support 9 is connected with the output shaft of the second stepping motor 8, and the second stepping motor 8 is fixed on the base 2 through another support vertical plate, and the output shaft of the second stepping motor 8 is perpendicular to the output shaft of the first stepping motor 6;

[0061] The system also contains through holes, 12 through holes with a diameter of 8mm are arranged on the shell of the loudspeaker array, there are 4 holes on the left, right and back respectively, for details, please refer to Figure 2 .

[0062] These through-holes are an application of precisely acoustically designed Helmholtz resonators, which primarily function to balance the air pressure inside and outside the cabinet and optimize the low-frequency radiation characteristics. Specifically, when the loudspeaker units (especially the low-frequency units) are working at high power, the dramatic vibration of the diaphragm can cause the air pressure inside the cabinet to change dramatically. These through-holes provide a controllable flow path for the air inside and outside the cabinet, effectively releasing excess pressure and reducing the constraints on the diaphragm movement, thereby reducing nonlinear distortion under large dynamic signals and improving transient response. At the same time, through careful design of the aperture and layout, it helps to disperse the high-order standing waves formed inside the cabinet and makes the low-frequency response smoother, improving the clarity and low-frequency efficiency of the sound. The through-hole design of this system is not isolated, it works in coordination with adaptive beamforming and mechanical steering: the optimized low-frequency response provides purer and lower-distortion original sound signals for electronic beamforming; balanced air pressure makes the loudspeaker units (especially the units responsible for low frequencies) work in a better linear range, which is crucial for maintaining sound quality consistency when the mechanical steering range is large and the dynamic beam tracking is performed.

[0063] This system realizes hybrid directivity control through the combination of electronic beamforming (main) and mechanical angle adjustment (auxiliary), with flexibility and high performance. At the same time, through multi-modal environmental perception (vision + acoustics) and intelligent algorithms, adaptive optimization in complex indoor environments is realized, significantly improving speech clarity and listening experience.

[0064] Embodiment 1:

[0065] This example provides a time-sharing directivity control method for a surround directivity loudspeaker system as described above, as shown in the accompanying Figure 7 The steps include:

[0066] Step 1: Initialize system configuration, set system working parameters; after system initialization, enter the continuous environmental perception loop, the key decision point is the judgment of the angle difference (Δθ);

[0067] Step 2: Environmental perception, collect environmental data: in the idle time of the loudspeaker, the sound intensity and phase difference indicators of each loudspeaker are perceived by the MEMS microphone in time-sharing mode, and the angle of the listener relative to the normal of the front of the sound box and the distance relative to the sound source and the 3D stereo figure of the sound field are detected in real time by the 3D laser radar;

[0068] Step 3: Parametric equalization calculation: the digital signal processor (DSP) uses a high-precision parametric equalizer to calibrate the data collected by the microphone in step 2 in the full frequency range of 20 Hz to 20 KHz. First, the frequency response curve is measured using the microphone, and the deviations exceeding ±2 dB are boosted or attenuated to compensate for the inherent deviations of the loudspeaker, solve the problem of uneven frequency response caused by standing waves and resonance, and compensate for the sound intensity difference caused by the authenticity or decay of different loudspeakers, so that the deviation is less than ±1 dB. The phase of different loudspeakers is calibrated to make the deviation less than 0.05 ms, as shown in the attached Figure 5 ;

[0069] Step 4: As shown in Figure 8 , Figure 9 , determine the direction of sound adjustment according to the signal source of different sound channels, such as straight ahead, left, right, left rear, right rear, etc. According to the 3D stereo figure of the listener, wall or reflecting surface obtained by the 3D laser radar, the path of sound from the sound box (center) through the wall or reflecting surface to the listener is calculated, which is specifically: 4a. Environment modeling and reflection point calculation: based on 3D environment data, the system establishes an acoustic reflection model. For left and right surround sound channels, the system calculates the best reflection point coordinates according to the listener position, left / right wall position and acoustic reflection law to ensure that the reflected sound energy reaches the listener accurately. Specifically, the sound source of the left channel is reflected by the left reflecting surface. According to the height of the sound source and the height of the listener, the distance from the sound source and the listener to the left reflecting surface determines the position of the reflection point to calculate the direction of emission. The right channel is calculated in the same way. For the rear surround sound channel, the system judges the reflection path: if the sound source is higher than the listener position and there is no obstruction in the middle, it is directly reflected by the back wall; otherwise, the system plans a path that reflects through the ceiling to the back wall, and then reflects to the listener through the back wall, and calculates the corresponding reflection point coordinates. These calculations ensure the accuracy of virtual sound source positioning;

[0070] 4b. Time-sharing scheduling: the system allocates a time slice to each virtual sound source direction (such as front, left, right, and rear). The time-sharing time interval is set to 5 ms, so that the total time of each direction is controlled within 35 ms (the upper limit of the Haas effect) once per cycle. This time allocation ensures that the human ear can integrate the sound in different time slices into a coherent sound field, while ensuring the response speed of the system to the change of the listener position;

[0071] If the sound pointing angle needs to be adjusted within 25°, electronic pointing compensation adjustment is performed: beamforming calculation is performed, beamforming parameters are calculated by DSP according to environmental data: time difference required for sound from each speaker unit to reach the listener is calculated according to listener position parameters including angle and distance, time compensation is performed on the electrical signal of each speaker unit through a delay circuit or a digital delay algorithm, phase difference of each speaker unit is calculated according to listener position and environmental parameters, based on the time difference and the phase difference, beamforming parameters required for each speaker unit are calculated, so that the sound waves reflected by different speakers through the target reflection point are superimposed and enhanced at the listener position, thereby realizing the directivity of the sound; if greater than 25°, step 6 is performed;

[0072] Specific sound pointing angle adjustment includes the following steps:

[0073] Step 4-1: DSP is used to calculate beamforming, beamforming parameters are calculated according to environmental data, and time difference required for sound from each speaker unit to reach the listener is calculated according to listener position (for example, x = 2.0 m, y = 1.4 m), wherein the speed constant c0 = 340 m / s, and it is assumed that the distance between two adjacent speaker units is d, and it is desired to direct the sound main beam to a direction with an angle of θ to the normal line of the array, then the path length difference (i.e. wave path difference) AL = d*sin(θ) exists in the sound paths from the two units to the listener position, in order to let the two sound waves be in phase and superimposed (enhanced) in the θ direction, the wave path difference AL must be compensated, and the wavelength λ of the sound wave is equal to the speed c / frequency f, and the time difference corresponding to the wave path difference AL is Δt = AL / c;

[0074] Step 4-2: FPGA needs to send a signal to the unit farther away from the listener in advance by Δt time, and the electrical signal of each speaker unit is compensated by a digital delay algorithm;

[0075] Step 4-3: the phase of the signal of the speaker unit farther away from the listener is ahead of a specific angle, i.e. phase difference, and the conversion relationship between them is: phase difference φ = (AL / λ)*360°, and the phase difference of each speaker unit is calculated according to listener position and environmental parameters;

[0076] Step 4-4: based on the time difference and the phase difference, beamforming parameters required for each speaker unit are calculated, and phase compensation values of each unit are obtained;

[0077] Step 5: Fine-tuning the directivity of the loudspeaker: according to the beamforming parameters, control the loudspeaker array, divide the audio signal into low-frequency signal and medium-high frequency signal through the frequency divider, deliver the low-frequency signal to the low-frequency array composed of 2 loudspeaker units, deliver the medium-high frequency signal to the medium-high frequency array composed of 6 loudspeaker units, realize real-time delay adjustment and phase adjustment of the medium-high frequency signal through FPGA high-speed parallel processing, the delay adjustment range is 0.1ms-2ms, and the phase adjustment range is -220°-220°; the target pointing angle of this step is the reflection point corresponding to the current time slice, not the listener position.

[0078] Step 6: If the sound pointing angle needs to be adjusted by more than 25°, first enable mechanical adjustment, and then superimpose electronic pointing compensation adjustment, wherein the mechanical adjustment is realized by calculating the rotation amount of the two stepper motor shafts, and then driving the stepper motor through the control unit to drive the angle adjustment mechanism to realize the mechanical direction adjustment of the box, realizing large range adjustment, which can at least meet the requirement of ±60° or more; the target pointing angle of this step is the reflection point corresponding to the current time slice, not the listener position.

[0079] Step 7: Return to step 2 and perform the next round of loop.

[0080] In step 6 of this example, the target of mechanical adjustment is within 5°, and in the process of mechanical movement, the binocular camera collects the listener angle in real time, and the control part judges at the same time that the sound pointing angle needs to be adjusted within 5°, if not, the mechanical part continues to rotate compensation, and the electronic pointing control continues to execute compensation, so the cycle is repeated until the angle is less than 5°, the mechanical compensation stops, and the angle compensated by the electronic pointing compensation adjustment at this time = the angle between the normal line of the sound box plane before the mechanical compensation is started and the line connecting the sound box and the listener - the angle adjusted by the mechanical compensation;

[0081] In step 6 of the example, if a large range of sound pointing angle adjustment is required while the listener moves quickly, the moving speed is greater than 1 m / s, and the response speed of the mechanical part cannot keep up with the quick movement of the listener, the electronic pointing is also superimposed on the mechanical movement, that is, during the mechanical rotation, the mechanical control part transmits the horizontal, i.e., left-right fluctuation and vertical (up-down) rotation angle to the beamforming control part in real time (the first stepping motor drive generates up-down movement to adapt to the height change of the listener caused by the distance change and standing and sitting postures; the second drive shaft generates left-right fluctuation to cooperate with the first drive shaft to generate pointing in any direction facing the listener); in the electronic pointing compensation adjustment, the beamforming control part calculates the difference between the angle of mechanical change and the angle of actual movement of the listener obtained by the binocular camera as the compensation angle of the electronic pointing, and performs fusion processing with the distance data of the listener, recalculates the beamforming parameters and adjusts the beamforming in real time to ensure that the sound is always accurately pointed to the listener. The mechanical adjustment stops, and the electronic pointing compensation operation continues to maintain, which can meet the occasion of quick movement of the listener and the need for large range adjustment of sound pointing angle. While the mechanical mechanism moves smoothly to achieve a large range of coverage, the electronic system responds at high speed to handle dynamic details, thereby realizing a tracking range and response speed far beyond the limits of a single technology without sacrificing sound quality.

[0082] Compared with the prior art, the present application has the following beneficial effects: 1. The present application realizes precise control of sound directivity by combining mechanical adjustment and electronic adjustment, overcoming the problem of inaccurate directivity of traditional loudspeaker arrays. The system can adjust the propagation direction and range of sound as needed, effectively solving the problem of poor listening experience caused by excessive or insufficient sound diffusion range; 2. The present application uses frequency division processing technology to process low-frequency and medium-high-frequency sound separately and optimize them respectively, realizing uniform frequency response. By setting multiple through holes in the shell of the loudspeaker device, the sound of the loudspeaker unit forms an anti-phase, effectively optimizing the directivity of the low-frequency sound. At the same time, using a medium-high-frequency loudspeaker array and beamforming technology, precise directivity control of medium-high-frequency sound is realized, effectively solving the problem of excessive concentration of medium-high-frequency sound or excessive diffusion of low-frequency sound; 3. The present application constructs a complete adaptive system that can automatically adjust directivity and frequency response according to environmental changes or listener position; through environmental perception and intelligent algorithm, the listener's position is monitored in real time, and the directivity of the loudspeaker array is dynamically adjusted as needed, greatly improving the adaptability and flexibility of the system, solving the problem of complex manual adjustment operation in the prior art; 4. The present application optimizes the processing of low-frequency and medium-high-frequency sound separately. Through low-frequency loudspeaker array and signal preprocessing technology, the effect of low-frequency sound is improved, at the same time, using medium-high-frequency enhancement technology, parametric equalization and beamforming principle, the performance of medium-high-frequency sound is significantly improved, effectively solving the problems of low-frequency sound turbidity and medium-high-frequency sound harshness; 5. The design scheme of the present application is simple and feasible, easy to implement, and can be applied to various scenes. Through frequency division processing and optimization design, the system structure is more compact, the cost is lower, the application range is expanded, and the limitation of complex directivity optimization method in the prior art is overcome.

Claims

1. A surround directional loudspeaker system, comprising a cuboid enclosure, characterized in that, The box top is provided with a 3D laser radar, and the box is connected with a mechanical steering mechanism. The 3D laser radar is installed on the top of the sound box, and is used for scanning and analyzing the angle θ of the listener relative to the central axis and the distance relative to the sound source, and obtaining the three-dimensional graph of the entire sound field. The mechanical steering mechanism is also connected with the control unit and is arranged outside the whole box body, and is used for adjusting the orientation of the whole box body, and is suitable for occasions where the sound pointing angle range needs to be adjusted to more than 25°; the loudspeaker unit array is composed of 8 loudspeaker units, wherein 2 low-frequency loudspeaker units arranged at both ends of the array are in the form of a coil loudspeaker; 6 middle-high-frequency loudspeaker units arranged in the middle and in a ring shape are in the form of a coil combined with a piezoelectric loudspeaker, the piezoelectric mode is suitable for occasions exceeding 10KHz, the diameter of the ring is 400±50mm, the diaphragm area of the 6 middle-high-frequency loudspeaker units is 150±20mm 2 , the acoustic cavity volume is 40±5cm 3 , and the sensitivity is 105±5dB; The frequency division and signal processing unit adopts a cooperative processing architecture of a digital signal processor (DSP) and a field programmable gate array (FPGA), the digital signal processor (DSP) is configured to perform digital frequency division processing, parametric equalization and beamforming of the audio signal, the frequency division point is obtained through the middle position of the frequency response curve of the mid-high frequency loudspeaker and the low frequency loudspeaker, and the low frequency signal below 500Hz±50Hz is transmitted to the two low frequency loudspeaker units, and the mid-high frequency signal above 500Hz±50Hz is transmitted to the six mid-high frequency loudspeaker units; the field programmable gate array (FPGA) is connected with the digital signal processor (DSP) to form digital pointing adjustment of the loudspeaker sound direction, receives the parametric equalization and beamforming coefficient instructions from the DSP, the parametric equalization is used for calibrating each loudspeaker, and the beamforming is used for real-time phase and delay adjustment of the signal transmitted to the six mid-high frequency loudspeaker units, so as to realize the directivity adjustment of less than 25° direction. The microphone adopts a MEMS microphone and is arranged at the center position of the six mid-high frequency loudspeaker units in the sound box, and is used for perceiving the direct sound emitted by each loudspeaker unit in different time, that is, the six loudspeakers emit sound at different times to calibrate the sound periodically; the input end of the control unit is connected with the environment perception unit including the MEMS microphone and the binocular camera, and the output end is connected with the digital signal processor (DSP) in the frequency division and signal processing unit, the control unit receives and fuses the data of the environment perception unit, dynamically calculates the parametric equalization and beamforming parameters required by each loudspeaker unit in the loudspeaker array based on the fused data, and sends the parameters to the digital signal processor.

2. The surround directional speaker system of claim 1, wherein, The shell is provided with twelve through holes with a diameter of 8±1mm, which are distributed on the left, right and rear sides of the shell, four on each side, for balancing the air pressure inside and outside the box and optimizing the low frequency radiation characteristics.

3. The surround directional speaker system of claim 1, wherein, The mechanical steering mechanism is provided with a sound box connecting piece and a base, the base is provided with a universal rotation unit, the upper surface of the universal rotation unit is fixedly connected with the bottom of the box body through the sound box connecting piece; the universal rotation unit is in a cylindrical shape, two pitch driving shafts arranged along the same radial direction are oppositely arranged on the side surface of the outer shell of the universal rotation unit, the open ends of the U-shaped supports are respectively connected with the two pitch driving shafts, the closed ends of the U-shaped supports are connected with the output shafts of the first stepper motors, the first stepper motors are fixed on the base through the support vertical plates; the outer shell side surface of the universal rotation unit is also provided with left and right heave driving shafts, the left and right heave driving shafts are arranged in a radial direction perpendicular to the pitch shafts, the left and right heave driving shafts are connected with the output shafts of the second stepper motors through V-shaped rotating supports, wherein, the V-shaped rotating supports are arranged in a V-shaped top-down manner, one end of the rotating support is connected with the output shaft of the second stepper motor, the second stepper motor is fixed on the base through another support vertical plate, and the output shaft of the second stepper motor is perpendicular to the output shaft of the first stepper motor.

4. A time division directional control method of a surround directional speaker system as claimed in any one of claims 1 to 3, characterized in that, The control unit is further configured to perform a time-sharing surround control method: based on the sound field three-dimensional figure obtained by the 3D laser radar, the spatial relationship between the listener and the main reflecting surface in the room is established; the audio signal is time-shared and distributed to different time slices according to the virtual sound source direction; in each time slice, the speaker unit array is controlled to point the sound wave main beam to the corresponding target reflecting surface, so as to form a surround sound field through reflected sound at the listener position; And the control unit is configured to perform the following angle adjustment cooperative control operation: when the angle between the listener position and the normal line of the front face of the sound box is greater than a first preset threshold, the mechanical steering mechanism is controlled to adjust the orientation of the box body, and at the same time, the frequency division and signal processing unit is controlled to perform electronic beam forming compensation on the speaker unit array; when the angle between the listener position and the normal line of the front face of the sound box is reduced to below a second preset threshold, the mechanical steering mechanism is stopped, and only the frequency division and signal processing unit is used for electronic beam forming tracking.

5. The time division steering control method of a surround directional speaker system according to claim 4, wherein, The method comprises the following steps: Step 1: initialize system configuration, set system working parameters; Step 2: environmental perception, collect environmental data: in the idle time of the loudspeaker, the sound intensity and phase difference indicators of each loudspeaker are perceived by the MEMS microphone in a time-sharing manner, the angle of the listener relative to the normal line of the front face of the sound box and the distance relative to the sound source are detected in real time by the 3D laser sensor, and the three-dimensional figure of the entire sound field is scanned at the same time; Step 3: parameter balance calculation: the digital signal processor (DSP) uses a high-precision parameter balancer to calibrate the data collected by the microphone in step 2 in the full frequency band, i.e. 20Hz-20KHz, first, the frequency response curve is measured by the microphone, and the deviation exceeding ±2dB is boosted or attenuated to compensate for the deviation of the loudspeaker itself, solve the problem of uneven frequency response caused by standing wave and resonance, compensate for the sound intensity difference caused by the authenticity or time decay of different loudspeakers to make the deviation less than ±1dB, and calibrate the phase of different loudspeakers to make the deviation less than 0.05ms. Step 4: According to the signal source of different sound channels, the direction of the sound needs to be adjusted. According to the 3D stereo figure of the listener, wall or reflecting surface obtained by the 3D laser radar, the path of the sound from the sound box through the wall or reflecting surface to the listener is calculated: 4a. Environment modeling and reflection point calculation: Based on the 3D environment data, the system establishes an acoustic reflection model. For the left and right surround sound channels, the system calculates the optimal reflection point coordinates according to the listener position, left / right wall position and acoustic reflection law, to ensure that the reflected sound energy reaches the listener accurately. The sound source of the left channel is reflected through the left reflecting surface. According to the height of the sound source and the height of the listener, the distance from the sound source and the listener to the left reflecting surface determines the position of the reflection point, thereby calculating the direction of emission. The right channel is calculated in the same way. For the rear surround sound channel, the system judges the reflection path: if the sound source is higher than the listener position and there is no obstruction in the middle, it is directly reflected through the back wall; otherwise, the system plans a path that reflects through the ceiling to the back wall, and then reflects to the listener through the back wall, and calculates the corresponding reflection point coordinates; 4b. Time-sharing scheduling: The system allocates a time slice to each virtual sound source direction. The time-sharing time interval is set to 5ms, so that the total time of each direction is controlled within 35ms and circulates once every 35ms; If the angle of the sound to be adjusted is within 25°, electronic pointing compensation adjustment is performed: beamforming calculation is performed, and beamforming parameters are calculated through DSP according to environmental data: according to the listener position parameters, including angle and distance, the time difference required for the sound of each loudspeaker unit to reach the listener is calculated, and the electrical signal of each loudspeaker unit is delayed through a delay circuit or a digital delay algorithm. According to the listener position and environmental parameters, the phase difference of each loudspeaker unit is calculated. Based on these time difference and phase difference, the beamforming parameters required for each loudspeaker unit are calculated, so that the sound waves of different loudspeakers are superimposed and enhanced at the listener position after reflection through the target reflection point, thereby realizing the directivity of the sound; if it is greater than 25°, step 6 is performed; Step 5: Directionality fine-tuning of loudspeakers: According to the beamforming parameters, the loudspeaker array is controlled, the audio signal is divided into low frequency signal and medium-high frequency signal through a frequency divider, the low frequency signal is delivered to a low frequency array composed of 2 loudspeaker units, and the medium-high frequency signal is delivered to a medium-high frequency array composed of 6 loudspeaker units. Through FPGA high-speed parallel processing, real-time delay adjustment and phase adjustment of medium-high frequency signal are realized, the delay adjustment range is 0.1ms-2ms, and the phase adjustment range is -220°-220°; Step 6: If the sound pointing angle needs to be adjusted beyond the range of 25°, mechanical adjustment is enabled first, and then electronic pointing compensation adjustment is superimposed. The mechanical adjustment is realized by calculating the rotation amount of the shaft of the two step motors, and then driving the step motor through the control unit to drive the angle adjustment mechanism to realize the mechanical direction adjustment of the box, realizing a large range of adjustment, which can at least meet the requirement of ±60° or more; Step 7: Return to step 2 and perform the next round of circulation.

6. The time division steering control method of a surround directional speaker system according to claim 5, wherein, In step 6, the target of mechanical adjustment is within 5°, during the mechanical movement, the 3D laser radar collects the listener angle in real time, and the control part judges whether the electronic pointing adjustment needs to be within 5° at the same time, if not, the mechanical part continues to rotate to compensate, and the electronic pointing control continues to compensate, so the cycle is repeated until the angle is less than 5°, the mechanical compensation stops, and the angle compensated by the electronic pointing compensation adjustment = the angle between the normal of the sound box plane before the mechanical compensation starts and the line connecting the sound box and the listener - the angle adjusted by the mechanical compensation.

7. The time division steering control method of a surround directional speaker system according to claim 6, wherein, In step 6, if a large range of sound pointing angle adjustment is needed and includes rapid movement of the listener, the movement speed is greater than 1m / s, the response speed of the mechanical part cannot keep up with the rapid movement of the listener, and electronic pointing needs to be superimposed during mechanical movement, that is, during the mechanical rotation, the mechanical control part transmits the horizontal, i.e. left and right fluctuation and vertical inclination rotation angle to the beam forming control part in real time, wherein the first stepper motor drive generates inclination action to adapt to the height change of the listener caused by the distance change of the listener and the standing and sitting posture; the second drive shaft generates left and right fluctuation action, which cooperates with the first drive shaft to generate pointing in any direction facing the listener; in the electronic pointing compensation adjustment, the beam forming control part calculates the difference between the angle changed by the mechanical part and the angle actually moved by the listener obtained by the binocular camera as the compensation angle of the electronic pointing, and fuses the distance data of the listener to re-calculate the beam forming parameters and adjust the beam forming in real time to ensure that the sound is always accurately pointed to the listener. The mechanical adjustment stops, and the electronic pointing compensation operation continues to maintain, which can meet the occasion of rapid movement of the listener and large range of sound pointing angle adjustment. While the mechanical mechanism moves smoothly to achieve a wide range of coverage, the electronic system responds at high speed to handle dynamic details, so as to realize the tracking range and response speed far beyond the single technology limit without sacrificing sound quality.

8. The method of claim 4, wherein, In step 4, the electronic pointing compensation adjustment is achieved by the following measures: step 4-1: using DSP to calculate beam forming, calculating beam forming parameters according to environmental data, and calculating the time difference required for sound from each speaker unit to reach the listener according to the position of the listener, wherein the speed constant c0 = 340m / s, assuming that the distance between two adjacent speaker units is d, and the target points the sound main beam to a direction with an angle θ with the array normal, then the path length difference between the sound from the two units to the listener position is ΔL = d*sin(θ), in order to let the two sound waves superimpose in phase in the θ direction, i.e. enhance, it is necessary to compensate for this wave path difference ΔL, the wavelength of the sound wave λ = sound speed c / frequency f, and the time difference corresponding to this wave path difference ΔL is Δt = ΔL / c; Step 4-2: FPGA needs to send signals to the unit far from the listener in advance by Δt time, and the digital delay algorithm is used to delay and compensate the electrical signals of each speaker unit. Step 4-3: The signals of the loudspeaker units far away from the listener are made to have a certain phase difference, i.e. a phase lead, and the conversion relationship therebetween is: phase difference φ=(ΔL / λ)*360°, and the phase difference of each loudspeaker unit is calculated according to the position of the listener and the environmental parameters; Step 4-4: Based on the time difference and the phase difference, the beamforming parameters required by each loudspeaker unit are calculated to obtain the phase compensation value of each unit.

Images