WO2019227279A1 - 降噪方法、装置和无人机 - Google Patents

降噪方法、装置和无人机 Download PDF

Info

Publication number
WO2019227279A1
WO2019227279A1 PCT/CN2018/088673 CN2018088673W WO2019227279A1 WO 2019227279 A1 WO2019227279 A1 WO 2019227279A1 CN 2018088673 W CN2018088673 W CN 2018088673W WO 2019227279 A1 WO2019227279 A1 WO 2019227279A1
Authority
WO
WIPO (PCT)
Prior art keywords
sound
drone
collected
flight
noise
Prior art date
Application number
PCT/CN2018/088673
Other languages
English (en)
French (fr)
Inventor
刘政哲
赵丛
封旭阳
李思晋
Original Assignee
深圳市大疆创新科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to PCT/CN2018/088673 priority Critical patent/WO2019227279A1/zh
Priority to CN201880031271.XA priority patent/CN110709921A/zh
Publication of WO2019227279A1 publication Critical patent/WO2019227279A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
    • G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L15/16—Speech classification or search using artificial neural networks

Definitions

  • the invention relates to the technical field of unmanned aerial vehicles, in particular to a noise reduction method, a device and an unmanned aerial vehicle.
  • the invention provides a noise reduction method, a device and an unmanned aerial vehicle, so as to reduce the noise generated by the unmanned aerial vehicle during flight, and further enables the unmanned aerial vehicle to obtain the real sound of the environment.
  • an embodiment of the present invention provides a noise reduction method, which is applied to a drone, and includes:
  • the characteristic parameters of the compensation sound are determined according to the characteristic parameters of the noise generated by the power system of the drone during flight;
  • the sound generating device is controlled to generate a compensation sound according to the characteristic parameter of the compensation sound to suppress the noise generated by the power system during the flight of the UAV.
  • an embodiment of the present invention provides a method for noise reduction of a sound collected by a drone, including:
  • Acquiring sounds collected by the drone during flight wherein the collected sounds include sounds generated by ambient sound sources and noise generated by the drone's power system during flight;
  • an embodiment of the present invention provides a drone, including: a memory, a processor, a power system, and a sound generating device;
  • the memory is used to store program code
  • the processor calls the program code, and when the program code is executed, is used to perform the following operations:
  • the characteristic parameters of the compensation sound are determined according to the characteristic parameters of the noise generated by the power system of the drone during flight;
  • the sound generating device is controlled to generate a compensation sound according to the characteristic parameter of the compensation sound to suppress the noise generated by the power system during the flight of the UAV.
  • an embodiment of the present invention provides a noise reduction device for reducing noise collected by a drone, including: a memory and a processor;
  • the memory is used to store program code
  • the processor calls the program code, and when the program code is executed, is used to perform the following operations:
  • Acquiring sounds collected by the drone during flight wherein the collected sounds include sounds generated by ambient sound sources and noise generated by the drone's power system during flight;
  • an embodiment of the present invention provides a readable storage medium, and the readable storage medium stores a computer program; when the computer program is executed, the first aspect or the second aspect of the embodiment of the present invention is provided. Noise reduction method.
  • the invention provides a noise reduction method, a device and a drone.
  • a characteristic parameter of a compensation sound is determined according to a characteristic parameter of a noise generated by a power system during flight of the drone, and a sound generating device is controlled to generate a sound according to the characteristic parameter of the compensation sound.
  • Compensation sound can interact with the noise generated by the power system to cancel or reduce the intensity of the noise, so as to achieve the effect of suppressing noise in real time, reduce the noise generated by the drone during flight, and improve the drone during flight.
  • the degree of environmental friendliness in China is the use of drones to capture the real sound of the environment.
  • FIG. 1 is a schematic architecture diagram of an unmanned flight system applicable to an embodiment of the present invention
  • FIG. 2 is a flowchart of a noise reduction method according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of a structure of an unmanned aerial vehicle according to an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of determining a frequency domain data component according to an embodiment of the present invention.
  • FIG. 5 is a flowchart of a method for noise reduction of a sound collected by a drone according to an embodiment of the present invention
  • FIG. 6 is a schematic structural diagram of an unmanned aerial vehicle provided by an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a noise reduction device according to an embodiment of the present invention.
  • a component when a component is called “fixed to” another component, it may be directly on another component or a centered component may exist. When a component is considered to be “connected” to another component, it can be directly connected to another component or a centered component may exist at the same time.
  • FIG. 1 is a schematic architecture diagram of an unmanned flight system applicable to an embodiment of the present invention.
  • this embodiment uses a drone as a rotary wing drone as an example for a schematic description.
  • the drone may also be a jet drone.
  • the unmanned aerial system 100 may include an unmanned aerial vehicle 110.
  • the drone 110 may include a power system 150, a flight control system 160, a rack, and a gimbal 120 carried on the rack.
  • the unmanned aerial system 100 may further include a control terminal 130.
  • the drone 110 may perform wireless communication with the control terminal 130.
  • the frame may include a fuselage and a tripod (also called a landing gear).
  • the fuselage may include a center frame and one or more arms connected to the center frame. One or more arms extend radially from the center frame.
  • the tripod is connected to the fuselage, and is used to support the UAV 110 when landing.
  • the power system 150 may include one or more electronic governors (referred to as ESCs) 151, one or more propellers 153, and one or more electric motors 152 corresponding to the one or more propellers 153, where the electric motor 152 is connected to Between the electronic governor 151 and the propeller 153, the motor 152 and the propeller 153 are arranged on the arm of the drone 110; the electronic governor 151 is used to receive the driving signal generated by the flight control system 160 and provide driving according to the driving signal Current is supplied to the motor 152 to control the rotation speed of the motor 152.
  • the motor 152 is used to drive the propeller to rotate, so as to provide power for the flight of the drone 110, and the power enables the drone 110 to achieve one or more degrees of freedom of movement.
  • the drone 110 may rotate about one or more rotation axes.
  • the rotation axis may include a roll axis (Roll), a yaw axis (Yaw), and a pitch axis (Pitch).
  • the motor 152 may be a DC motor or an AC motor.
  • the motor 152 may be a brushless motor or a brushed motor.
  • the flight control system 160 may include a flight controller 161 and a sensing system 162.
  • the sensing system 162 is used to measure the attitude information of the drone, that is, the position information and status information of the drone 110 in space, such as three-dimensional position, three-dimensional angle, three-dimensional velocity, three-dimensional acceleration, and three-dimensional angular velocity.
  • the sensing system 162 may include, for example, at least one of a gyroscope, an ultrasonic sensor, an electronic compass, an Inertial Measurement Unit (IMU), a vision sensor, a global navigation satellite system, and a barometer.
  • the global navigation satellite system may be a Global Positioning System (GPS).
  • GPS Global Positioning System
  • the flight controller 161 is used to control the flight of the drone 110.
  • the flight controller 161 may control the flight of the drone 110 according to the attitude information measured by the sensing system 162. It should be understood that the flight controller 161 may control the drone 110 according to a pre-programmed program instruction, and may also control the drone 110 by taking a picture.
  • the gimbal 120 may include a motor 122.
  • the gimbal is used to carry the photographing device 123.
  • the flight controller 161 may control the movement of the gimbal 120 through the motor 122.
  • the PTZ 120 may further include a controller for controlling the movement of the PTZ 120 by controlling the motor 122.
  • the gimbal 120 may be independent of the drone 110 or may be a part of the drone 110.
  • the motor 122 may be a DC motor or an AC motor.
  • the motor 122 may be a brushless motor or a brushed motor.
  • the gimbal can be located on top of the drone or on the bottom of the drone.
  • the photographing device 123 may be, for example, a device for capturing an image, such as a camera or a video camera.
  • the photographing device 123 may communicate with the flight controller and perform shooting under the control of the flight controller to obtain photos and / or videos.
  • the flight controller may also The drone 110 is controlled based on an image captured by the photographing device 123.
  • the photographing device 123 of this embodiment includes at least a photosensitive element, such as a complementary metal oxide semiconductor (Complementary Metal Oxide Semiconductor) sensor or a charge-coupled device (CCD) sensor. It can be understood that the shooting device 123 can also be directly fixed on the drone 110, so that the PTZ 120 can be omitted.
  • a photosensitive element such as a complementary metal oxide semiconductor (Complementary Metal Oxide Semiconductor) sensor or a charge-coupled device (CCD) sensor.
  • CCD charge-coupled device
  • the control terminal 130 is located on the ground side of the unmanned flight system 100 and can communicate with the drone 110 wirelessly. In some embodiments, the control terminal 130 can send control instructions to the drone 110 wirelessly. Control the drone to perform corresponding actions, such as flying actions, shooting actions, etc. In some embodiments, the control terminal 130 includes a display device, and the display device may be used to display attitude information of the drone 110. In addition, the image captured by the imaging device may also be displayed on the control terminal 130. It should be understood that the control terminal 130 may be a device independent of the drone 110.
  • the power system 150 may cause a poor environmental friendliness of the drone during the flight.
  • drones are generally not equipped with a sound collection device for capturing sound.
  • the shooting device 123 obtains photos or videos without sounds. When the photos or videos are played back later, The background sound of the photo or video is configured, which will cause the real sound in the environment where the drone 110 is located to be lost, and the real shooting scene cannot be restored.
  • a sound collection device (not shown) is configured on the control terminal 130, and the sound collection device configured on the control terminal can collect the real sound of the environment in which the control terminal is located.
  • the drone 110 and the control The terminal 130 may be in different scenarios.
  • the distance between the drone 110 and the control terminal 130 is relatively long.
  • the sound collected by the sound collection device configured on the control terminal 130 may be different from the environment in which the drone 110 is located. There is a large deviation in the real sound, which cannot restore the real shooting scene.
  • the drone 110 may include a sound collection device 170, where the sound collection device 170 may be any sensor that collects ambient sounds, such as a microphone, etc., where the sound collection device 170 may For one or more, the sound collection device 170 may be disposed outside the rack, or may be disposed in the rack.
  • the sound collection device can collect the sound of the environment in which the drone 110 is located during the flight of the drone 110.
  • the power system 150 generates a large noise during the flight of the drone 110, the sound collection device is caused 170 cannot collect the real sound of the environment in which the drone 110 is located, that is, the sound generated by the ambient sound source, wherein the ambient sound source can be any sound source in the environment except the drone.
  • the drone 110 may further include a sound generating device 180.
  • the sound generating device 180 may be any device capable of receiving a control signal and generating a sound according to the control signal, such as an audio generator, a speaker, or the like.
  • the drone 110 may control the sound generating device 180 to generate a compensation sound, wherein the compensation sound interacts with the noise generated by the power system 150 to reduce or suppress the production of the power system 150 In this way, the noise generated by the power system 150 is suppressed.
  • the sound collection device 170 can collect the real sound of the environment in which the drone 110 is located, and restore the real shooting scene. The noise reduction method provided by the embodiment of the present invention will be described in detail below.
  • FIG. 2 is a flowchart of a noise reduction method according to an embodiment of the present invention. As shown in FIG. 2, the noise reduction method provided in this embodiment may be applied to a drone, and the noise reduction method may include:
  • the execution subject of the method provided by the embodiment of the present invention is a drone, and further, it may be a processor of the drone, wherein the processor may be a processor in a flight controller as described above.
  • the processor may be a processor other than a flight controller, and the processor may be one or more, which work individually or cooperatively to perform the method of the embodiment of the present invention.
  • the processor of the drone may obtain the characteristic parameters of the compensation sound, wherein the characteristic parameters of the compensation sound are based on the characteristics of the noise generated by the power system of the drone during the flight.
  • the parameters are determined, that is, the signal characteristics of the compensation sound are determined according to the signal characteristics of the noise generated by the power system.
  • the characteristic parameter may include at least one of frequency, phase, and amplitude.
  • the signal characteristics of the compensation sound are determined according to the signal characteristics of the noise generated by the power system and include: the frequency of the compensation sound is the same as the frequency of the noise, and the phase of the compensation sound is opposite to the phase of the noise.
  • the signal characteristics of the compensation sound are determined according to the signal characteristics of the noise generated by the power system and include: the phase of the compensation sound is opposite to that of the noise, and the amplitude of the compensation sound is the same as the amplitude of the noise.
  • the signal characteristics of the compensation sound are determined according to the signal characteristics of the noise generated by the power system and include: the frequency of the compensation sound is the same as the frequency of the noise, and the phase of the compensation sound is opposite to that of the noise, and the compensation sound Has the same amplitude as the noise.
  • the processor of the drone may control the sound generating device provided on the drone to generate the compensation sound in real time according to the acquired characteristic parameters of the compensation sound. For example, in some cases, the processor may generate a control signal according to a characteristic parameter of the compensation sound, and send the control signal to a sound generating device, and the sound generating device generates a compensation sound according to the control signal. In some cases, the processor may send characteristic parameters of the compensation sound to the sound generating device, and the sound generating device may generate a corresponding compensation sound according to the received characteristic parameters.
  • the noise generated by the power system cancels the sound with the compensation sound generated by the sound generating equipment, reduces the intensity of the noise generated by the power system, and effectively suppresses the noise.
  • the compensation sound generated by the sound generating device can completely eliminate the noise generated by the power system .
  • the invention provides a noise reduction method.
  • the characteristic parameters of the compensation sound are determined according to the characteristic parameters of the noise generated by the power system during the flight of the drone, and the sound generating device is controlled to generate the compensation sound according to the characteristic parameters of the compensation sound. Interact with the noise generated by the power system.
  • the noise is cancelled with the compensation sound to cancel or reduce the intensity of the noise, so as to achieve the effect of suppressing the noise in real time, reduce the noise generated by the drone during flight, and improve the drone's flight.
  • the degree of environmental friendliness in the process is the use of drones to capture the real sound of the environment.
  • the number of sound generating devices and the installation positions of the sound generating devices on the drone are not limited, and may be set according to the number and positions of the propellers included in the power system.
  • the number of sound generating devices may be one or more, and the number of sound generating devices is the same as the number of propellers included in the power system.
  • the noise generated by each propeller corresponds to a compensation sound, and each compensation sound is generated and played by a sound generating device.
  • a sound generating device corresponding to each propeller may be installed on a corresponding propeller shaft. For example, FIG.
  • the power system includes four propellers (the propellers 11 to 14 respectively), and the drone further includes four sound generating devices 16.
  • the sound generating device 16 may be provided on a rotating shaft of the propeller.
  • the acquiring the characteristic parameters of the compensation sound includes: acquiring the characteristic parameters of the compensation sound from a storage device configured by the drone.
  • the noise generated by the power system during the flight of the drone is collected in a quiet experimental environment, and the characteristic parameters of the noise are determined.
  • the characteristic parameters of the compensation sound stored in the storage device are based on the characteristic parameters of the noise. definite.
  • a storage device is configured on the drone, and the storage device can store characteristic parameters of the compensation sound.
  • the storage device may be a local storage device provided inside the drone.
  • the characteristic parameters of the compensation sound are stored in the storage device in advance.
  • the processor of the drone can obtain the characteristic parameters of the compensation sound from the storage device, and then control based on the obtained characteristic parameters of the compensation sound.
  • the sound generating device generates a corresponding compensation sound.
  • control terminal stores the characteristic parameters of the compensation sound
  • the UAV may obtain the characteristic parameters of the compensation sound through a wireless link with the control terminal.
  • the noise reduction method provided in this embodiment may further include: determining a flying state of the drone.
  • obtaining the characteristic parameters of the compensation sound from the storage device configured by the drone may include: obtaining the characteristic parameters of the compensation sound corresponding to the flight status from the storage device configured by the drone.
  • the noise generated by the power system may be different depending on the flight status of the drone, which results in different characteristics of the noise generated by the power system under different flight conditions.
  • the characteristic parameters of multiple sets of compensation sounds stored in the storage device are pre-stored, wherein each of the characteristic parameters of the multiple sets of compensation sounds It is determined according to the characteristic parameters of the noise generated by the power system of the UAV under each different flight state.
  • the drone's processor can determine the current flight status in real time. After determining the drone's flight status, it can use the flight status in the feature parameters stored in the storage device in advance. A characteristic parameter of the compensation sound corresponding to the flight state is determined.
  • the flight state may include one or more of an accelerated flight state, a decelerated flight state, a hovering state, a turning state, an ascending flight state, and a descending flight state.
  • the processor may obtain the characteristic parameters of the compensation sound corresponding to the hovering state from the storage device, and then control the sound generating device to generate the compensation sound according to the obtained characteristic parameters. Suppress the noise generated by the power system when the drone is in hovering state.
  • the accuracy and flexibility of determining the compensation sound is improved.
  • obtaining the characteristic parameters of the compensated sound includes: obtaining sounds collected by a sound collection device configured by the drone, wherein the sounds collected by the sound collection device include the drone During the flight, the noise generated by the power system and the sound generated by the ambient sound source are used to determine the characteristic parameters of the noise generated by the power system in the collected sound, and the characteristic parameters of the compensation sound are determined according to the characteristic parameters of the noise generated by the power system.
  • the drone is equipped with a sound collection device.
  • the sound collection device collects sound, wherein the collected sound includes the ambient sound source in the environment in which it is located.
  • the processor of the drone acquires the sound collected by the sound collection device, analyzes the collected sound, determines the characteristic parameters of the noise generated by the power system in the collected sound, and can further determine according to the characteristic parameters of the noise Compensate the characteristic parameters of the sound.
  • the sound collection device may be a microphone array.
  • the microphone array may be an existing integrated device.
  • the microphone array may be formed of multiple devices capable of collecting sound, and the types of multiple devices capable of collecting sound may be the same or different.
  • determining the characteristic parameters of the noise generated by the power system in the collected sound may include: performing frequency domain transformation on the sound collected by the sound collection device to obtain frequency domain data of the collected sound, and according to the frequency domain data Determine the characteristic parameters of the noise generated by the power system.
  • the sound collected by the sound collection device is a time-domain signal
  • the sound collected by the sound collection device can be subjected to a fast Fourier transform (Fast Fourier Transformation, FFT) and other frequency domain transformations to obtain frequency domain data.
  • the domain data may include a phase spectrum and / or an amplitude spectrum obtained by frequency domain transformation, and the processor may determine the characteristic parameters of the noise generated by the power system according to the acquired frequency domain data.
  • determining the characteristic parameter of the noise generated by the power system according to the frequency domain data includes: determining frequency domain data corresponding to the noise from the collected frequency domain data of the sound, and according to the The frequency-domain data corresponding to the noise determines a characteristic parameter of the noise generated by the power system.
  • the frequency-domain data obtained after the frequency-domain transformation includes frequency-domain data of sounds generated by ambient sound sources and frequency-domain data of noise generated by the power system of the drone.
  • the drone processor may obtain the frequency data from The frequency domain data corresponding to the noise generated by the power system is determined in the method, and the characteristic parameters of the noise generated by the power system are determined according to the frequency domain data corresponding to the noise generated by the power system.
  • acquiring the sound collected by the sound collection device configured by the drone may include: obtaining sound collected by two sound collection devices; and collecting the sound collection device Acquire the frequency domain data of the acquired sound by performing frequency domain transformation, including: performing frequency domain transformation on the sound collected by each of the two sound acquisition equipments to obtain the frequency domain of the two sets of sounds acquired Data; determining the frequency-domain data corresponding to the noise from the collected frequency-domain data of the sound includes: determining the noise in the spectral data of the two sets of sound according to the spectral data of the two sets of sound and the installation position of the power system on the drone Frequency domain data components.
  • multiple sound collection devices can be set on the drone, such as at least two sound collection devices, and the multiple sound collection devices can collect sounds at the same time.
  • the processor of the drone may obtain the sounds collected by the sound collection device A and the sound collection device B, respectively, where the sound collection device A and The sound collected by the sound collection device B includes the noise generated by the rotation of the propeller C.
  • the processor performs frequency domain transformation on the sound collected by the sound collection device A and the sound collection device B to obtain the frequency of the sound collected by the sound collection device A.
  • the installation position on the machine determines the frequency domain data of the sound collected by the sound collection device A and the frequency domain data of the sound collected by the sound collection device B.
  • Frequency domain data propeller noise generated C The following will explain in detail how to determine the frequency domain data components of the noise in the spectral data of the two sets of sounds based on the spectral data of the two sets of sounds and the installation position of the power system on the drone.
  • the processor may divide the frequency-domain data of each set of sounds into multiple frequency-domain data components in units of frequencies, that is, multiple frequencies of the sounds collected by the sound collection device A
  • the domain data component corresponds to multiple frequency domain data components of the sound collected by the sound collection device B, that is, the frequency of each frequency domain data component of the sound collected by the sound collection device A is collected by the corresponding sound collection device B.
  • the frequency of the frequency-domain data component of the sound is the same, where the multiple frequency-domain data components include the frequency-domain data component of the noise generated by the ambient sound source and the frequency-domain data component of the noise generated by the power system.
  • the processor of the drone can determine the power according to the frequency domain data component of the sound collected by the sound collection device A, the frequency domain data component of the sound collected by the corresponding sound collection device B, and the installation position of the power system on the drone.
  • Frequency-domain data components of system-generated noise The following will describe in detail the determination of the power system generated based on the frequency domain data components of the sound collected by the sound collection device A, the frequency domain data components of the sound collected by the corresponding sound collection device B, and the installation position of the power system on the drone. Specific principles of frequency-domain data components of noise.
  • a frequency domain data component of the sound collected by the sound collection device A and a frequency domain data component of the sound collected by the corresponding sound collection device B are selected. It is assumed that the two frequency domain data components correspond to each other.
  • this sound source may be an ambient sound source or a power system, such as the propeller C. If the distance from the sound source to the sound collection device A is s1, and the distance from the sound source to the sound collection device B is s2, according to the two frequency domain data components, it can be determined that the sound recording device A and the sound collection device B record The sound phase difference ⁇ ⁇ , the distance difference between this sound source and the sound collection device A and the sound collection device B can be determined according to the following formula:
  • the position of this sound source is necessarily on the hyperbola C1, C2 determined by
  • the processor can determine whether the power system (propeller C) is on the hyperbola according to the installation position of the power system on the drone. When the power system (propeller C) is not on the hyperbola, the sound source can be identified as an ambient sound source. When the power system (propeller C) is on the hyperbola, it can be determined that the sound source is most likely to be a power system. In some embodiments, when the power system (propeller C) is on the hyperbola, the sound source can be identified. For the power system.
  • the method described above can be used to determine multiple sets of hyperboloids, and the position of the sound source must be at the intersection of the multiple sets of hyperboloids. At this time, you can According to the installation position of the power system on the drone, it is determined whether the power system (propeller C) is at the intersection of multiple sets of hyperboloids, so that the judgment accuracy can be greatly improved.
  • FIG. 5 is a flowchart of a method for noise reduction of a sound collected by a drone according to an embodiment of the present invention.
  • the noise reduction method provided in this embodiment may be implemented by a noise reduction device, where a drone or a terminal device includes the noise reduction device.
  • the terminal device may include one or more of a remote controller, a smart phone, a tablet computer, a laptop computer, a desktop computer, and a wearable device (such as a watch or a wristband, etc.).
  • the terminal device may include the foregoing Control terminal.
  • the method for reducing noise collected by a drone provided in this embodiment may include:
  • the drone may be equipped with a sound collection device.
  • the sound collection device collects sound during the drone flight, and the collected sound includes the sound generated by the ambient sound source. Sound and noise generated by the drone's power system during flight.
  • a processor of the noise reduction device may obtain a sound collected by a sound collection device.
  • the noise reduction device may be wired or wirelessly connected to the drone to obtain the sound collected by the sound collection device, and the noise reduction device may be The human-machine establishes a communication connection directly or indirectly to obtain the sound collected by the sound collection device.
  • S502 Input the collected sound into a neural network model to obtain a noise after noise reduction.
  • the noise reduction device may have a built-in neural network model built in, and the noise reduction device may input the collected sound into the trained neural network model, where the neural network model is used to eliminate the The noise generated by the power system of the drone during flight, the neural network model can output the sound produced by the ambient sound source, that is, the real sound of the environment where the drone is located.
  • the method for noise reduction of the sound collected by the drone provided in this embodiment can eliminate the noise generated by the power system in the sound collected by the drone through the neural network model. In this way, the drone can be obtained. The real sound of the environment.
  • neural networks include, but are not limited to, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and long short-term memory networks (LSTMs).
  • CNNs convolutional neural networks
  • RNNs recurrent neural networks
  • LSTMs long short-term memory networks
  • the neural network model takes as output the sounds generated by the ambient sound sources collected in multiple different scenes, and uses the sounds generated by the ambient sound sources collected in multiple different scenes and
  • the mixed sound of the noise generated by the power system of the drone during the flight is obtained by input training.
  • the neural network requires a large number of data samples for training.
  • sounds generated by environmental sound sources collected in multiple different scenarios can be used as output, that is, acquired in multiple different scenarios. (Such as quiet indoor scenes, road scenes, square scenes, woods scenes, etc.) the real sound of the environment, using the real sound as the output of the neural network network; during training, the environment collected in multiple different scenes
  • a mix of sound generated by the sound source and noise generated by the drone's power system during flight is used as input.
  • a feasible way is to collect the noise generated by the power system of the drone during the flight in a quiet experimental environment, and collect the noise in multiple different scenarios.
  • the sound generated by the ambient sound source is fused to obtain the mixed sound.
  • Another feasible way is that the drone flies in the plurality of different scenes, and collects it through a sound collection device during the flight. Sound, at this time, the collected sound is the mixed sound.
  • the neural network model is trained by using multiple sets of inputs and corresponding outputs. After the training is completed, the neural network model can be used to eliminate noise generated by the power system during the flight of the collected sound.
  • the noise generated by the power system of the UAV during flight includes noise generated by the power system corresponding to multiple flight states of the UAV during flight.
  • the multiple flight states may include at least two of an accelerated flight state, a decelerated flight state, a hovering state, a steering state, an ascending flight state, and a descending flight state.
  • the noise generated by the power system may be different depending on the flight status of the drone. It is possible to acquire noise generated by the power system in multiple flight states to train the neural network, which can effectively improve the noise reduction performance of the neural network model.
  • the obtained neural network model is more accurate.
  • the method for noise reduction provided by the UAV in this embodiment may be combined with the noise reduction method provided by the embodiments shown in FIG. 2 to FIG. 4 described above.
  • the sound collected by the drone is obtained.
  • the sound collected by the drone may be the sound obtained after the drone executes S201 to S202, that is, the sound collected by the sound collection device.
  • Input the neural network model to obtain the noise after the noise reduction. In this way, the noise collected by the sound collection device can be further reduced by data processing.
  • FIG. 6 is a schematic structural diagram of an unmanned aerial vehicle according to an embodiment of the present invention.
  • the drone provided by this embodiment may execute the noise reduction method provided by the method embodiments shown in FIG. 2 to FIG. 4.
  • the drone provided in this embodiment may include: a memory 62, a processor 61, a power system 63, and a sound generating device (not shown).
  • the memory 62 is configured to store a program code.
  • the processor 61 calls program code, and when the program code is executed, is used to perform the following operations:
  • the characteristic parameters of the compensation sound are determined according to the characteristic parameters of the noise generated by the power system 63 during the flight of the UAV.
  • the sound generating device is controlled to generate the compensation sound according to the characteristic parameter of the compensation sound to suppress the noise generated by the power system 63 during the flight of the UAV.
  • the characteristic parameter includes at least one of frequency, phase, and amplitude.
  • the processor 61 is specifically configured to:
  • the characteristic parameters of the compensation sound are obtained from the storage device configured by the drone.
  • the processor 61 is further configured to:
  • the processor 61 is specifically configured to:
  • the characteristic parameters of the compensation sound corresponding to the flight status are obtained from a storage device configured by the drone.
  • the flight state includes one or more of an accelerated flight state, a decelerated flight state, a hovering state, a steering state, an ascending flight state, and a descending flight state.
  • the drone may further include a sound collection device, and the processor 61 is specifically configured to:
  • the sound collected by the sound collection device includes the noise generated by the power system 63 during the flight of the drone and the sound generated by the ambient sound source.
  • the characteristic parameters of the noise generated by the power system 63 in the collected sound are determined.
  • the characteristic parameters of the compensation sound are determined according to the characteristic parameters of the noise generated by the power system 63.
  • the processor 61 is specifically configured to:
  • the characteristic parameters of the noise generated by the power system 63 are determined according to the frequency domain data.
  • the processor 61 is specifically configured to:
  • the frequency domain data corresponding to the noise is determined from the frequency domain data of the collected sound.
  • the characteristic parameters of the noise generated by the power system 63 are determined according to the frequency-domain data corresponding to the noise.
  • the processor 61 is specifically configured to:
  • Frequency-domain transform is performed on the sounds collected by each of the two sound collection devices to obtain the frequency-domain data of the two sets of collected sounds.
  • the frequency domain data component corresponding to the noise in the spectral data of the two sets of sound is determined according to the spectral data of the two sets of sounds and the installation position of the power system 63 on the drone.
  • the sound collection device is a microphone array.
  • the frequency of the compensation sound is the same as the frequency of the noise, and the phase of the compensation sound is opposite to that of the noise.
  • the frequency of the compensation sound is the same as the frequency of the noise, and the phase of the compensation sound is opposite to that of the noise, and the amplitude of the compensation sound is the same as the amplitude of the noise.
  • the unmanned aerial vehicle provided by the embodiment of the present invention is configured to execute the noise reduction method provided by the method embodiments shown in FIG. 2 to FIG. 4 of the present invention.
  • the technical principles and technical effects are similar, and are not described herein again.
  • FIG. 7 is a schematic structural diagram of a noise reduction device according to an embodiment of the present invention.
  • the noise reduction device provided by this embodiment may execute the method for noise reduction of a sound collected by a drone provided by the method embodiment shown in FIG. 5.
  • the noise reduction device provided in this embodiment is configured to reduce noise of a sound collected by a drone, and may include a memory 72 and a processor 71.
  • the memory 72 is configured to store a program code.
  • the processor 71 calls program code, and when the program code is executed, is used to perform the following operations:
  • the collected sound includes the sound generated by the ambient sound source and the noise generated by the drone's power system during the flight.
  • the collected sound is input to a neural network model to obtain a noise-reduced sound.
  • the neural network model is used to eliminate the noise generated by the power system during the flight of the collected sound.
  • the neural network model takes the sound generated by the ambient sound sources collected in multiple different scenes as the output, the sound generated by the ambient sound sources collected in multiple different scenes, and the power of the drone during flight.
  • the noise generated by the system is obtained from the input training.
  • the noise generated by the power system of the UAV during flight includes noise generated by the power system corresponding to multiple flight states of the UAV during flight.
  • the multiple flight states include at least two of an accelerated flight state, a decelerated flight state, a hovering state, a steering state, an ascent flight state, and a descending flight state.
  • the noise reduction device provided by the embodiment of the present invention is configured to execute the method for noise reduction of a sound collected by a drone provided by the method embodiment shown in FIG. 5 of the present invention.
  • the technical principles and technical effects are similar. More details.
  • a person of ordinary skill in the art may understand that all or part of the steps of implementing the foregoing method embodiments may be implemented by a program instructing related hardware.
  • the aforementioned program may be stored in a computer-readable storage medium.
  • the steps including the foregoing method embodiments are performed; and the foregoing storage medium includes various media that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Soundproofing, Sound Blocking, And Sound Damping (AREA)

Abstract

一种降噪方法、装置和无人机。降噪方法应用于无人机（110），包括：获取补偿声音的特征参数；其中，补偿声音的特征参数是根据无人机（110）在飞行过程中动力系统（150）产生的噪声的特征参数确定的；根据补偿声音的特征参数控制声音发生设备（180）产生补偿声音以抑制无人机（110）在飞行过程中动力系统（150）产生的噪声。该降噪方法可以有效地降低无人机（110）在飞行过程中产生的噪声。

Description

降噪方法、装置和无人机 技术领域
本发明涉及无人机技术领域，尤其涉及一种降噪方法、装置和无人机。
背景技术
随着无人机技术的发展，越来越多的用户开始使用无人机执行航拍、勘测、巡检等任务。无人机在飞行时，螺旋桨的转速很高，螺旋桨与空气摩擦会产生很大的噪声，导致无人机在飞行过程中产生较大的噪声。另外，由于无人机在飞行过程中产生的噪声，导致无人机无法获取无人机所处环境的真实声音。
发明内容
本发明提供一种降噪方法、装置和无人机，以降低无人机在飞行过程中产生的噪声，进一步地，使得无人机可以获取所处环境的真实声音。
第一方面，本发明实施例提供一种降噪方法，应用于无人机，包括：
获取补偿声音的特征参数；其中，所述补偿声音的特征参数是根据无人机在飞行过程中动力系统产生的噪声的特征参数确定的；
根据所述补偿声音的特征参数控制声音发生设备产生补偿声音以抑制所述无人机在飞行过程中所述动力系统产生的噪声。
第二方面，本发明实施例提供一种用于对无人机采集的声音进行降噪的方法，包括：
获取无人机在飞行过程中采集到的声音，其中，所述采集到的声音包括环境声源产生的声音和无人机在飞行过程中无人机的动力系统产生的噪声；
将所述采集到的声音输入神经网络模型以获取降噪后的声音；其中，所述神经网络模型用于消除所述采集到的声音中所述无人机在飞行过程中动力系统产生的噪声。
第三方面，本发明实施例提供一种无人机，包括：存储器、处理器、动力系统和声音发生设备；
所述存储器，用于存储程序代码；
所述处理器，调用所述程序代码，当所述程序代码被执行时，用于执行以下操作：
获取补偿声音的特征参数；其中，所述补偿声音的特征参数是根据无人机在飞行过程中动力系统产生的噪声的特征参数确定的；
根据所述补偿声音的特征参数控制声音发生设备产生补偿声音以抑制所述无人机在飞行过程中所述动力系统产生的噪声。
第四方面，本发明实施例提供一种降噪装置，用于对无人机采集的声音进行降噪，包括：存储器和处理器；
所述存储器，用于存储程序代码；
所述处理器，调用所述程序代码，当所述程序代码被执行时，用于执行以下操作：
获取无人机在飞行过程中采集到的声音，其中，所述采集到的声音包括环境声源产生的声音和无人机在飞行过程中无人机的动力系统产生的噪声；
将所述采集到的声音输入神经网络模型以获取降噪后的声音；其中，所述神经网络模型用于消除所述采集到的声音中所述无人机在飞行过程中动力系统产生的噪声。
第五方面，本发明实施例提供一种可读存储介质，所述可读存储介质上存储有计算机程序；所述计算机程序在被执行时，实现本发明实施例第一方面或者第二方面提供的降噪方法。
本发明提供一种降噪方法、装置和无人机，根据无人机飞行过程中动力系统产生的噪声的特征参数确定补偿声音的特征参数，根据所述补偿声音的特征参数控制声音发生设备生成补偿声音，补偿声音可以与动力系统产生的噪声相互作用以抵消或者减弱噪声的强度，从而达到实时抑制噪声的效果，降低无人机在飞行过程中产生的噪声，提高了无人机在飞行过程中的环境友好程度，有利用无人机采集所处环境的真实声音。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下 面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1为本发明实施例适用的无人飞行系统的示意性架构图；
图2为本发明实施例提供的降噪方法的流程图；
图3为本发明实施例提供的无人机的一种结构的结构示意图；
图4为本发明实施例提供的确定频域数据分量的原理示意图；
图5为本发明实施例提供的用于对无人机采集的声音进行降噪的方法的流程图；
图6为本发明实施例提供的无人机的结构示意图；
图7为本发明实施例提供的降噪装置的结构示意图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
需要说明的是，当组件被称为“固定于”另一个组件，它可以直接在另一个组件上或者也可以存在居中的组件。当一个组件被认为是“连接”另一个组件，它可以是直接连接到另一个组件或者可能同时存在居中组件。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本发明。本文所使用的术语“及/或”包括一个或多个相关的所列项目的任意的和所有的组合。在不冲突的前提下，具体实施方式的各个实施例可以相互组合。
图1为本发明实施例适用的无人飞行系统的示意性架构图。其中，本实施例以无人机为旋翼无人机为例进行示意性说明，在其他实施例中，所述无人机也可以为喷气式无人机。
无人飞行系统100可以包括无人机110。无人机110可以包括动力系统 150、飞行控制系统160、机架和承载在机架上的云台120。可选的，无人飞行系统100还可以包括控制终端130。无人机110可以与控制终端130进行无线通信。
机架可以包括机身和脚架(也称为起落架)。机身可以包括中心架以及与中心架连接的一个或多个机臂，一个或多个机臂呈辐射状从中心架延伸出。脚架与机身连接，用于在无人机110着陆时起支撑作用。
动力系统150可以包括一个或多个电子调速器(简称为电调)151、一个或多个螺旋桨153以及与一个或多个螺旋桨153相对应的一个或多个电机152，其中电机152连接在电子调速器151与螺旋桨153之间，电机152和螺旋桨153设置在无人机110的机臂上；电子调速器151用于接收飞行控制系统160产生的驱动信号，并根据驱动信号提供驱动电流给电机152，以控制电机152的转速。电机152用于驱动螺旋桨旋转，从而为无人机110的飞行提供动力，该动力使得无人机110能够实现一个或多个自由度的运动。在某些实施例中，无人机110可以围绕一个或多个旋转轴旋转。例如，上述旋转轴可以包括横滚轴(Roll)、偏航轴(Yaw)和俯仰轴(pitch)。应理解，电机152可以是直流电机，也可以交流电机。另外，电机152可以是无刷电机，也可以是有刷电机。
飞行控制系统160可以包括飞行控制器161和传感系统162。传感系统162用于测量无人机的姿态信息，即无人机110在空间的位置信息和状态信息，例如，三维位置、三维角度、三维速度、三维加速度和三维角速度等。传感系统162例如可以包括陀螺仪、超声传感器、电子罗盘、惯性测量单元(Inertial Measurement Unit，IMU)、视觉传感器、全球导航卫星系统和气压计等传感器中的至少一种。例如，全球导航卫星系统可以是全球定位系统(Global Positioning System，GPS)。飞行控制器161用于控制无人机110的飞行，例如，可以根据传感系统162测量的姿态信息控制无人机110的飞行。应理解，飞行控制器161可以按照预先编好的程序指令对无人机110进行控制，也可以通过拍摄画面对无人机110进行控制。
云台120可以包括电机122。云台用于携带拍摄装置123。飞行控制器161可以通过电机122控制云台120的运动。可选地，作为另一实施例，云台120还可以包括控制器，用于通过控制电机122来控制云台120的运动。 应理解，云台120可以独立于无人机110，也可以为无人机110的一部分。应理解，电机122可以是直流电机，也可以是交流电机。另外，电机122可以是无刷电机，也可以是有刷电机。还应理解，云台可以位于无人机的顶部，也可以位于无人机的底部。
拍摄装置123例如可以是照相机或摄像机等用于捕获图像的设备，拍摄装置123可以与飞行控制器通信，并在飞行控制器的控制下进行拍摄以获取照片和/或视频，飞行控制器也可以根据拍摄装置123拍摄的图像控制无人机110。本实施例的拍摄装置123至少包括感光元件，该感光元件例如为互补金属氧化物半导体(Complementary Metal Oxide Semiconductor，CMOS)传感器或电荷耦合元件(Charge-coupled Device，CCD)传感器。可以理解，拍摄装置123也可直接固定于无人机110上，从而云台120可以省略。
控制终端130位于无人飞行系统100的地面端，可以通过无线方式与无人机110进行通信，在某些实施例中，所述控制终端130可以通过无线方式向无人机110发送控制指令来控制无人机执行相应的动作，例如飞行动作、拍摄动作等。在某些实施例中，控制终端130包括显示设备，所述显示设备可以用于显示无人机110的姿态信息。另外，还可以在控制终端130上显示拍摄装置拍摄的图像。应理解，控制终端130可以是独立于无人机110的设备。
目前，由于无人机110在飞行过程中，动力系统150会产生较大的噪声，这样会导致无人机在飞行过程中的环境友好性较差。另外，在某些情况中，无人机一般不配置用于采集声音的声音采集设备，拍摄装置123获取不带声音的照片或者视频，后期在播放所述照片或者视频的时候，再重新为所述照片或者视频配置背景声音，这样会导致丢失无人机110所处环境中的真实声音，无法还原真实的拍摄场景。在某些情况中，所述控制终端130上配置声音采集设备(未示出)，配置在控制终端上的声音采集设备可以采集控制终端所处环境的真实声音，然而，无人机110和控制终端130可能处于不同的场景中，例如，无人机110与控制终端130之间的距离较远，配置在控制终端130上的声音采集设备采集到的声音可能与无人机110所处环境的真实声音存在较大偏差，无法还原真实的拍摄场景。
在本发明实施例中，所述无人机110可以包括声音采集设备170，其中， 所述声音采集设备170可以为任何采集环境声音的传感器，例如麦克风等，其中，所述声音采集设备170可以为一个或多个，所述声音采集设备170可以设置在机架外，也可以设置在机架内。所述声音采集设备可以在无人机110飞行的过程中采集无人机110所处的环境的声音，然而，由于无人机110在飞行过程中动力系统150产生较大噪声，导致声音采集设备170不能采集到无人机110所处的环境的真实声音，即环境声源产生的声音，其中环境声源可以为环境中除无人机以外任何声源。因此，所述无人机110中还可以包括声音发生设备180，其中，声音发生设备180可以为任何能够接收控制信号并根据控制信号来生成声音的设备，例如音频发生器、喇叭等设备，在无人机110飞行的过程中，无人机110可以控制所述声音发生设备180产生补偿声音，其中，所述补偿声音和所述动力系统150产生的噪声相互作用以降低或抑制动力系统150产生的噪声，这样由于动力系统150产生的噪声被抑制，在拍摄装置123进行拍摄时，所述声音采集设备170可以采集到无人机110所处环境的真实声音，还原了真实的拍摄场景。下面将对本发明实施例提供的降噪方法进行详细地说明。
图2为本发明实施例提供的降噪方法的流程图。如图2所示，本实施例提供的降噪方法，可以应用于无人机，降噪方法可以包括：
S201、获取补偿声音的特征参数。
具体地，本发明实施例提供的方法的执行主体为无人机，进一步地，可以为无人机的处理器，其中，所述处理器可以为如前所述的飞行控制器中的处理器，在某些情况中，所述处理器可以为飞行控制器之外的处理器，所述处理器可以为一个或多个，单独地或协同地工作以执行本发明实施例的方法。
其中，在无人机飞行的过程中，无人机的处理器可以获取补偿声音的特征参数，其中，所述补偿声音的特征参数是根据无人机在飞行过程中动力系统产生的噪声的特征参数确定的，即补偿声音的信号特征是根据动力系统产生的噪声的信号特征来确定的。可选的，特征参数可以包括频率、相位、振幅中的至少一种。
可选的，所述补偿声音的信号特征是根据动力系统产生的噪声的信号特征来确定的包括：补偿声音的频率与噪声的频率相同，且补偿声音的相位与噪声的相位相反。
可选的，所述补偿声音的信号特征是根据动力系统产生的噪声的信号特征来确定的包括：补偿声音的相位与噪声的相位相反，且补偿声音的振幅与噪声的振幅相同。
可选的，所述补偿声音的信号特征是根据动力系统产生的噪声的信号特征来确定的包括：补偿声音的频率与噪声的频率相同，且补偿声音的相位与噪声的相位相反，且补偿声音的振幅与噪声的振幅相同。
S202、根据补偿声音的特征参数控制声音发生设备产生补偿声音以抑制无人机在飞行过程中动力系统产生的噪声。
具体的，在无人机飞行的过程中，无人机的处理器可以根据获取到的补偿声音的特征参数控制无人机上设置的声音发生设备实时产生补偿声音。例如，在某些情况中，所述处理器可以根据所述补偿声音的特征参数来产生控制信号，并利用所述控制信号发送给声音发生设备，声音发生设备根据所述控制信号产生补偿声音。在某些情况中，所述处理器可以将补偿声音的特征参数发送给声音发生设备，声音发生设备可以根据接收到的特征参数产生对应的补偿声音。这样，在无人机飞行过程中，动力系统产生的噪声与声音发生设备产生的补偿声音进行声音相消，降低了动力系统产生的噪声的强度，对所述噪声进行了有效的抑制。例如，当补偿声音的频率与噪声的频率相同，且补偿声音的相位与噪声的相位相反，且补偿声音的振幅与噪声的振幅相同，声音发生设备产生的补偿声音可以完全消除动力系统产生的噪声。
本发明提供一种降噪方法，根据无人机飞行过程中动力系统产生的噪声的特征参数确定补偿声音的特征参数，根据所述补偿声音的特征参数控制声音发生设备生成补偿声音，补偿声音可以与动力系统产生的噪声相互作用，噪声与补偿声音相消以抵消或者减弱噪声的强度，从而达到实时抑制噪声的效果，降低无人机在飞行过程中产生的噪声，提高了无人机在飞行过程中的环境友好程度，有利用无人机采集所处环境的真实声音。
需要说明的是，本实施例对于声音发生设备的数量和声音发生设备在无人机上的安装位置不做限定，可以根据动力系统中包括的螺旋桨的数量和位置进行设置。可选的，声音发生设备的数量可以为一个或者多个，声音发生设备的数量与动力系统包括的螺旋桨的数量相同。此时，每个螺旋桨产生的噪声均对应一个补偿声音，每个补偿声音分别由一个声音发生设备产生和播 放。可选的，为了提升噪声与补偿声音的声音相消效果，与每个螺旋桨分别对应的声音发生设备可以安装在对应螺旋桨的转轴上。示例性的，图3为本发明实施例提供的无人机的一种结构的结构示意图。如图3所示，动力系统包括4个螺旋桨(分别为螺旋桨11～螺旋桨14)，无人机还包括4个声音发生设备16。其中，声音发生设备16可以设置在螺旋桨的转轴上。
可选的，在一种实现方式中，所述获取补偿声音的特征参数包括：从无人机配置的存储装置中获取补偿声音的特征参数。
具体的，在安静的实验环境中采集无人机飞行过程中动力系统产生的噪声，并确定所述噪声的特征参数，存储在存储装置中的补偿声音的特征参数是根据所述噪声的特征参数确定的。无人机上配置有存储装置，所述存储装置可以存储所述补偿声音的特征参数。存储装置可以为设置在无人机内部的本地存储装置。存储装置中预先存储有补偿声音的特征参数，当无人机在飞行过程中，无人机的处理器可以从存储装置获取补偿声音的特征参数，然后再根据获取得到的补偿声音的特征参数控制声音发生设备产生对应的补偿声音。
可选地，在某些情况中，控制终端存储所述补偿声音的特征参数，无人机可以通过与控制终端之间的无线链路获取补偿声音的特征参数。
可选的，本实施例提供的降噪方法，还可以包括：确定无人机的飞行状态。相应的，从无人机配置的存储装置中获取补偿声音的特征参数，可以包括：从无人机配置的存储装置中获取与飞行状态对应的补偿声音的特征参数。
具体的，无人机的飞行状态不同，动力系统产生的噪声可能不同，这样导致无人机在不同飞行状态下，动力系统产生的噪声的特征参数也不相同。这样为了对不同飞行状态下无人机的动力系统产生的噪声精确地抑制，存储装置中预先存储的多组补偿声音的特征参数，其中，所述多组补偿声音的特征参数中的每一组是根据对应的每一个不同的飞行状态下无人机的动力系统产生的噪声的特征参数确定的。在无人机飞行的过程中，无人机的处理器可以实时地确定当前的飞行状态，在确定无人机的飞行状态后，可以根据所述飞行状态在存储装置中预先存储的特征参数中确定与所述飞行状态对应的补偿声音的特征参数。
可选的，飞行状态可以包括：加速飞行状态、减速飞行状态、悬停状态、 转向状态、上升飞行状态和下降飞行状态中的一种或多种。例如，当处理器确定无人机处于悬停状态时，处理器可以从存储装置中获取与悬停状态对应的补偿声音的特征参数，然后根据获取得到的特征参数控制声音发生设备产生补偿声音以抑制无人机处于悬停状态下时动力系统产生的噪声。通过确定无人机的飞行状态，进而根据飞行状态确定补偿声音的特征参数，提升了确定补偿声音的准确性和灵活性。
可选的，在另一种实现方式中，所述获取补偿声音的特征参数包括：获取无人机配置的声音采集设备采集到的声音，其中，声音采集设备采集到的声音包括无人机在飞行过程中动力系统产生的噪声和环境声源产生的声音，确定采集到的声音中动力系统产生的噪声的特征参数，根据动力系统产生的噪声的特征参数确定补偿声音的特征参数。
具体的，如前所述，无人机上配置有声音采集设备，在无人机在飞行的过程中，声音采集设备采集声音，其中，所述采集到的声音包括所处的环境中环境声源产生的声音和无人机的动力系统产生的噪声。无人机的处理器获取声音采集设备采集到的声音，对所述采集到的声音进行分析，确定所述采集到的声音中动力系统产生的噪声的特征参数，进而可以根据噪声的特征参数确定补偿声音的特征参数。通过获取无人机飞行过程中声音采集设备采集到的声音，并根据获取的声音确定补偿声音的特征参数，提升了确定补偿声音的特征参数的准确性和实时性。
可选的，声音采集设备可以为麦克风阵列。可选的，所述麦克风阵列可以为现有的集成设备。可选的，所述麦克风阵列可以由多个可以采集声音的设备形成，多个可以采集声音的设备的类型可以相同，也可以不同。
可选的，所述确定采集到的声音中动力系统产生的噪声的特征参数，可以包括：对声音采集设备采集到的声音进行频域变换获取采集到的声音的频域数据，根据频域数据确定动力系统产生的噪声的特征参数。
具体的，声音采集设备采集到的声音为时域信号，可以对声音采集设备采集到的声音进行快速傅立叶变换(Fast Fourier Transformation，FFT)等频域变换以获取频域数据，其中，所述频域数据可以包括频域变换得到的相位谱和/或幅值谱，处理器可以根据所述获取的频域数据来确定动力系统产生的噪声的特征参数。
进一步地，所述根据所述频域数据确定所述动力系统产生的噪声的特征参数，包括：从所述采集到的声音的频域数据中确定所述噪声对应的频域数据，根据所述噪声对应的频域数据确定所述动力系统产生的噪声的特征参数。
具体地，频域变换后获取的频域数据包括环境声源产生的声音的频域数据和无人机的动力系统产生的噪声的频域数据，无人机的处理器可以从获取的频率数据中确定动力系统产生的噪声对应的频域数据，并根据动力系统产生的噪声对应的频域数据来确定动力系统产生的噪声的特征参数。
可选的，本实施例提供的降噪方法，所述获取无人机配置的声音采集设备采集到的声音，可以包括：获取两个声音采集设备采集到的声音；所述对声音采集设备采集到的声音进行频域变换获取采集到的声音的频域数据，包括：对两个声音采集设备中每一个声音采集设备采集到的声音进行频域变换，获取采集到的两组声音的频域数据；所述从采集到的声音的频域数据中确定噪声对应的频域数据，包括：根据两组声音的频谱数据和动力系统在无人机上的安装位置确定两组声音的频谱数据中噪声的频域数据分量。
具体的，如前所述，无人机上可以设置多个声音采集设备，例如至少两个声音采集设备，所述多个声音采集设备可以同时采集声音，下面将以两个声音采集设备来进行示意性说明，如图4所示，在无人机飞行的过程中，无人机的处理器可以分别获取声音采集设备A和声音采集设备B采集到的声音，其中，所述声音采集设备A和声音采集设备B采集到的声音中包括螺旋桨C转动产生的噪声，处理器分别对声音采集设备A和声音采集设备B采集到的声音进行频域变换以获取声音采集设备A采集到的声音的频域数据和声音采集设备B采集到的声音的频域数据，处理器根据声音采集设备A采集到的声音的频域数据、声音采集设备B采集到的声音的频域数据和螺旋桨C在无人机上的安装位置确定声音采集设备A采集到的声音的频域数据和声音采集设备B采集到的声音的频域数据中螺旋桨C产生的噪声的频域数据。下面将详细解释如何根据两组声音的频谱数据和动力系统在无人机上的安装位置确定两组声音的频谱数据中噪声的频域数据分量。
在获取到两组声音的频域数据之后，处理器可以以频率为单位将每一组声音的频域数据划分成多个频域数据分量，即声音采集设备A采集到的声音的多个频域数据分量与声音采集设备B采集到的声音的多个频域数据分量一 一对应，即声音采集设备A采集到的声音的每一个频域数据分量的频率与对应的声音采集设备B采集到的声音的频域数据分量的频率相同，其中，多个频域数据分量中包括环境声源产生的噪声的频域数据分量和动力系统产生的噪声的频域数据分量。无人机的处理器可以根据声音采集设备A采集到的声音的频域数据分量、与对应的声音采集设备B采集到的声音的频域数据分量和动力系统在无人机上的安装位置确定动力系统产生的噪声的频域数据分量。下面将详细介绍根据声音采集设备A采集到的声音的频域数据分量、与其对应的声音采集设备B采集到的声音的频域数据分量和动力系统在无人机上的安装位置确定动力系统产生的噪声的频域数据分量的具体原理。
这里可以示意性说明，选择声音采集设备A采集到的声音的一个频域数据分量和与其对应的声音采集设备B采集到的声音的一个频域数据分量，假设这两个频域数据分量都对应于一个声源，这个声源可能为环境声源，也可能为动力系统，例如螺旋桨C。若这个声源到声音采集设备A的距离为s1，这个声源到声音采集设备B的距离为s2，根据这两个频域数据分量可以确定根据声音采集设备A和声音采集设备B记录到的声音相位差△φ，这个声源到声音采集设备A和声音采集设备B之间的距离差可以根据以下公式确定：
Figure PCTCN2018088673-appb-000001
其中，这两个频域数据分量对应的频率，声音在空气中传播速度为v＝340m/s。
由此可知，这个声源的位置必然在以|s1-s2|确定的双曲面C1、C2上，其中，双曲面C1、C2的焦点为声音采集设备A和声音采集设备B的位置。处理器可以根据动力系统在无人机上的安装位置判定动力系统(螺旋桨C)是否在这个双曲面上，当动力系统(螺旋桨C)不在这个双曲面上，可以认定这个声源为环境声源，当动力系统(螺旋桨C)在这个双曲面上，可以认定这个声源极有可能为动力系统，在某些实施例中，当动力系统(螺旋桨C)在这个双曲面上，可以认定这个声源为动力系统。可以理解的是，当存在3个或者更多的声音采集设备时，采用如前所述的方法可以确定多组双曲面，声源的位置必然在多组双曲面的交点上，此时，可以根据动力系统在无人机上的安装位置判定动力系统(螺旋桨C)是否在多组双曲面的交点上，这样判断精度可以大大提升。
图5为本发明实施例提供的用于对无人机采集的声音进行降噪的方法的流程图。本实施例提供的降噪方法，执行主体可以为降噪装置，其中，无人机或者终端设备包括所述降噪装置。终端设备可以包括遥控器、智能手机、平板电脑、膝上型电脑、台式电脑、穿戴式设备(例如手表或者手环等)中的一种或多种，所述终端设备可以包括如前所述的控制终端。如图5所示，本实施例提供的用于对无人机采集的声音进行降噪的方法，可以包括：
S501、获取无人机在飞行过程中采集到的声音。
具体地，如前所述，无人机可以配置声音采集设备，在无人机飞行过程中，声音采集设备在无人机飞行过程中采集声音，所述采集到的声音包括环境声源产生的声音和无人机在飞行过程中无人机的动力系统产生的噪声。当所述方法的执行主体为无人机包括的降噪装置时，降噪装置的处理器可以获取声音采集设备采集到的声音。当所述方法的执行主体为终端设备包括的降噪装置时，所述降噪装置可以与无人机有线或者无线地连接以获取声音采集设备采集到的声音，另外降噪装置可以通过与无人机直接或者间接的方式建立通信连接以获取声音采集设备采集到的声音。
S502、将采集到的声音输入神经网络模型以获取降噪后的声音。
具体地，降噪装置可以内置已经训练好的神经网络模型，所述降噪装置可以将采集到的声音输入到已经训练好的神经网络模型中，其中，所述神经网络模型用于消除采集到的声音中无人机在飞行过程中动力系统产生的噪声，神经网络模型即可以输出环境声源产生的声音，即无人机所处环境的真实声音。
本实施例提供的用于对无人机采集的声音进行降噪的方法，通过神经网络模型可以消除无人机采集的声音中动力系统产生的噪声，通过这种方式可以获取无人机所处环境的真实声音。
需要说明的是，本实施例不限定神经网络模型的类型。例如，神经网络包括但不限于卷积神经网络(convolutional neural network，CNN)，循环神经网络(recurrent neural network，RNN)以及长短期记忆网络(long short term memory，LSTM)。
可选的，在本实施例中，神经网络模型是以多个不同场景下采集到的环境声源产生的声音为输出、以多个不同场景下采集到的所述环境声源产生的 声音和无人机在飞行过程中动力系统产生的噪声的混合声音为输入训练获取的。具体地，所述神经网络需要大量的数据样本来进行训练，其中，在训练时，可以以多个不同场景下采集到的环境声源产生的声音为输出，即获取在多个不同的场景下(例如安静的室内场景、马路场景、广场场景、树林场景等)环境的真实声音，将所述真实声音作为神经网网络的输出；在训练时，以多个不同场景下采集到的所述环境声源产生的声音和无人机在飞行过程中动力系统产生的噪声的混合声音为输入。其中，混合声音获取的方式可以有如下两种：一种可行的方式，在安静的实验环境，采集无人机在飞行过程中动力系统产生的噪声，将所述噪声与多个不同场景下采集到的所述环境声源产生的声音融合以获取所述混合声音；另一种可行的方式，无人机在所述多个不同的场景下飞行，在飞行的过程中，通过声音采集设备采集声音，此时，所述采集到的声音即为所述混合声音。利用多组输入和对应的输出对神经网络模型进行训练，训练完成之后，所述神经网络模型即可以用于消除采集到的声音中无人机在飞行过程中动力系统产生的噪声。
可选的，在本实施例中，无人机在飞行过程中动力系统产生的噪声包括无人机在飞行过程中多个飞行状态对应的动力系统产生的噪声。其中，多个飞行状态可以包括：加速飞行状态、减速飞行状态、悬停状态、转向状态、上升飞行状态和下降飞行状态中的至少两个。
具体的，无人机的飞行状态不同，动力系统产生的噪声可能不同。可以获取多个飞行状态下所述动力系统产生的噪声来对所述神经网络进行训练，这样可以有效地提高神经网络模型的降噪性能。
通过采集无人机在飞行过程中多个飞行状态对应的动力系统产生的噪声，进而根据不同飞行状态对应的动力系统产生的噪声进行神经网络模型训练，使得获得的神经网络模型更加准确。
需要说明的是，本实施例提供的用于对无人机采集的声音进行降噪的方法，可以与上述图2～图4所示实施例提供的降噪方法相结合。
具体的，S501中，获取无人机采集到的声音，其中，所述无人机采集到的声音可以为无人机执行S201～S202之后获取的声音，即，将声音采集设备采集到的声音输入所述神经网络模型获取降噪后的声音，通过这种方式，可以通过数据处理的方式对声音采集设备采集到的声音进行进一步的降噪。
图6为本发明实施例提供的无人机的结构示意图。本实施例提供的无人机，可以执行图2～图4所示方法实施例提供的降噪方法。如图6所示，本实施例提供的无人机，可以包括：存储器62、处理器61、动力系统63和声音发生设备(未示出)。
存储器62，用于存储程序代码。
处理器61，调用程序代码，当程序代码被执行时，用于执行以下操作：
获取补偿声音的特征参数。其中，补偿声音的特征参数是根据无人机在飞行过程中动力系统63产生的噪声的特征参数确定的。
根据补偿声音的特征参数控制声音发生设备产生补偿声音以抑制无人机在飞行过程中动力系统63产生的噪声。
可选的，特征参数包括频率、相位、振幅中的至少一种。
可选的，处理器61具体用于：
从无人机配置的存储装置中获取补偿声音的特征参数。
可选的，处理器61还用于：
确定无人机的飞行状态。
处理器61具体用于：
从无人机配置的存储装置中获取与飞行状态对应的补偿声音的特征参数。
可选的，飞行状态包括：加速飞行状态、减速飞行状态、悬停状态、转向状态、上升飞行状态和下降飞行状态中的一种或多种。
可选的，无人机还可以包括声音采集设备，处理器61具体用于：
获取无人机配置的声音采集设备采集到的声音。其中，声音采集设备采集到的声音包括无人机在飞行过程中动力系统63产生的噪声和环境声源产生的声音。
确定采集到的声音中动力系统63产生的噪声的特征参数。
根据动力系统63产生的噪声的特征参数确定补偿声音的特征参数。
可选的，处理器61具体用于：
对声音采集设备采集到的声音进行频域变换获取采集到的声音的频域数据。
根据频域数据确定动力系统63产生的噪声的特征参数。
可选的，处理器61具体用于：
从采集到的声音的频域数据中确定噪声对应的频域数据。
根据噪声对应的频域数据确定动力系统63产生的噪声的特征参数。
可选的，处理器61具体用于：
获取两个声音采集设备采集到的声音。
对两个声音采集设备中每一个声音采集设备采集到的声音进行频域变换，获取采集到的两组声音的频域数据。
根据两组声音的频谱数据和动力系统63在无人机上的安装位置确定两组声音的频谱数据中噪声对应的频域数据分量。
可选的，声音采集设备为麦克风阵列。
可选的，补偿声音的频率与噪声的频率相同，且补偿声音的相位与噪声的相位相反。
可选的，补偿声音的频率与噪声的频率相同，且补偿声音的相位与噪声的相位相反，且补偿声音的振幅与噪声的振幅相同。
本发明实施例提供的无人机，用于执行本发明图2～图4所示方法实施例提供的降噪方法，其技术原理和技术效果类似，此处不再赘述。
图7为本发明实施例提供的降噪装置的结构示意图。本实施例提供的降噪装置，可以执行图5所示方法实施例提供的用于对无人机采集的声音进行降噪的方法。如图7所示，本实施例提供的降噪装置，用于对无人机采集的声音进行降噪，可以包括：存储器72和处理器71。
存储器72，用于存储程序代码。
处理器71，调用程序代码，当程序代码被执行时，用于执行以下操作：
获取无人机在飞行过程中采集到的声音。其中，采集到的声音包括环境声源产生的声音和无人机在飞行过程中无人机的动力系统产生的噪声。
将采集到的声音输入神经网络模型以获取降噪后的声音。其中，神经网络模型用于消除采集到的声音中无人机在飞行过程中动力系统产生的噪声。
可选的，神经网络模型是以多个不同场景下采集到的环境声源产生的声音为输出、以多个不同场景下采集到的环境声源产生的声音和无人机在飞行过程中动力系统产生的噪声的混合声音为输入训练获取的。
可选的，无人机在飞行过程中动力系统产生的噪声包括无人机在飞行过程中多个飞行状态对应的动力系统产生的噪声。
可选的，多个飞行状态包括：加速飞行状态、减速飞行状态、悬停状态、转向状态、上升飞行状态和下降飞行状态中的至少两个。
本发明实施例提供的降噪装置，用于执行本发明图5所示方法实施例提供的用于对无人机采集的声音进行降噪的方法，其技术原理和技术效果类似，此处不再赘述。
本领域普通技术人员可以理解：实现上述各方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成。前述的程序可以存储于一计算机可读取存储介质中。该程序在执行时，执行包括上述各方法实施例的步骤；而前述的存储介质包括：ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上各实施例仅用以说明本发明实施例的技术方案，而非对其限制；尽管参照前述各实施例对本发明实施例进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明实施例技术方案的范围。

Claims (32)

  1. 一种降噪方法，应用于无人机，其特征在于，包括：
    获取补偿声音的特征参数；其中，所述补偿声音的特征参数是根据无人机在飞行过程中动力系统产生的噪声的特征参数确定的；
    根据所述补偿声音的特征参数控制声音发生设备产生补偿声音以抑制所述无人机在飞行过程中所述动力系统产生的噪声。
  2. 根据权利要求1所述的方法，其特征在于，所述特征参数包括频率、相位、振幅中的至少一种。
  3. 根据权利要求1或2所述的方法，其特征在于，所述获取补偿声音的特征参数，包括：
    从所述无人机配置的存储装置中获取所述补偿声音的特征参数。
  4. 根据权利要求3所述的方法，其特征在于，所述方法还包括：
    确定所述无人机的飞行状态；
    所述从所述无人机配置的存储装置中获取所述补偿声音的特征参数，包括：
    从所述无人机配置的所述存储装置中获取与所述飞行状态对应的补偿声音的特征参数。
  5. 根据权利要求4所述的方法，其特征在于，所述飞行状态包括：加速飞行状态、减速飞行状态、悬停状态、转向状态、上升飞行状态和下降飞行状态中的一种或多种。
  6. 根据权利要求1或2所述的方法，其特征在于，所述获取补偿声音的特征参数，包括：
    获取所述无人机配置的声音采集设备采集到的声音；其中，所述声音采集设备采集到的声音包括所述无人机在飞行过程中所述动力系统产生的噪声和环境声源产生的声音；
    确定所述采集到的声音中所述动力系统产生的噪声的特征参数；
    根据所述动力系统产生的噪声的特征参数确定所述补偿声音的特征参数。
  7. 根据权利要求6所述的方法，其特征在于，所述确定所述采集到的声音中所述动力系统产生的噪声的特征参数，包括：
    对所述声音采集设备采集到的声音进行频域变换获取所述采集到的声音 的频域数据；
    根据所述频域数据确定所述动力系统产生的噪声的特征参数。
  8. 根据权利要求7所述的方法，其特征在于，所述根据所述频域数据确定所述动力系统产生的噪声的特征参数，包括：
    从所述采集到的声音的频域数据中确定所述噪声对应的频域数据；
    根据所述噪声对应的频域数据确定所述动力系统产生的噪声的特征参数。
  9. 根据权利要求8所述的方法，其特征在于，
    所述获取所述无人机配置的声音采集设备采集到的声音，包括：
    获取两个声音采集设备采集到的声音；
    所述对所述声音采集设备采集到的声音进行频域变换获取所述采集到的声音的频域数据，包括：
    对所述两个声音采集设备中每一个声音采集设备采集到的声音进行频域变换，获取采集到的两组声音的频域数据；
    所述从所述采集到的声音的频域数据中确定所述噪声对应的频域数据，包括：
    根据所述两组声音的频谱数据和所述动力系统在所述无人机上的安装位置确定所述两组声音的频谱数据中所述噪声对应的频域数据分量。
  10. 根据权利要求9所述的方法，其特征在于，所述声音采集设备为麦克风阵列。
  11. 根据权利要求2所述的方法，其特征在于，所述补偿声音的频率与所述噪声的频率相同，且所述补偿声音的相位与所述噪声的相位相反。
  12. 根据权利要求2或11所述的方法，其特征在于，所述补偿声音的频率与所述噪声的频率相同，且所述补偿声音的相位与所述噪声的相位相反，且所述补偿声音的振幅与所述噪声的振幅相同。
  13. 一种用于对无人机采集的声音进行降噪的方法，其特征在于，包括：
    获取无人机在飞行过程中采集到的声音，其中，所述采集到的声音包括环境声源产生的声音和无人机在飞行过程中无人机的动力系统产生的噪声；
    将所述采集到的声音输入神经网络模型以获取降噪后的声音；其中，所述神经网络模型用于消除所述采集到的声音中所述动力系统产生的噪声。
  14. 根据权利要求13所述的方法，其特征在于，所述神经网络模型是以 多个不同场景下采集到的环境声源产生的声音为输出、以所述多个不同场景下采集到的所述环境声源产生的声音和无人机在飞行过程中动力系统产生的噪声的混合声音为输入训练获取的。
  15. 根据权利要求14所述的方法，其特征在于，所述无人机在飞行过程中动力系统产生的噪声包括无人机在飞行过程中多个飞行状态对应的动力系统产生的噪声。
  16. 根据权利要求15所述的方法，其特征在于，所述多个飞行状态包括：加速飞行状态、减速飞行状态、悬停状态、转向状态、上升飞行状态和下降飞行状态中的至少两个。
  17. 一种无人机，其特征在于，包括：存储器、处理器、动力系统和声音发生设备；
    所述存储器，用于存储程序代码；
    所述处理器，调用所述程序代码，当所述程序代码被执行时，用于执行以下操作：
    获取补偿声音的特征参数；其中，所述补偿声音的特征参数是根据无人机在飞行过程中动力系统产生的噪声的特征参数确定的；
    根据所述补偿声音的特征参数控制声音发生设备产生补偿声音以抑制所述无人机在飞行过程中所述动力系统产生的噪声。
  18. 根据权利要求17所述的无人机，其特征在于，所述特征参数包括频率、相位、振幅中的至少一种。
  19. 根据权利要求17或18所述的无人机，其特征在于，所述处理器具体用于：
    从所述无人机配置的存储装置中获取所述补偿声音的特征参数。
  20. 根据权利要求19所述的无人机，其特征在于，所述处理器还用于：
    确定所述无人机的飞行状态；
    所述处理器具体用于：
    从所述无人机配置的所述存储装置中获取与所述飞行状态对应的补偿声音的特征参数。
  21. 根据权利要求20所述的无人机，其特征在于，所述飞行状态包括：加速飞行状态、减速飞行状态、悬停状态、转向状态、上升飞行状态和下降 飞行状态中的一种或多种。
  22. 根据权利要求17或18所述的无人机，其特征在于，所述无人机还包括声音采集设备，所述处理器具体用于：
    获取所述声音采集设备采集到的声音；其中，所述声音采集设备采集到的声音包括所述无人机在飞行过程中所述动力系统产生的噪声和环境声源产生的声音；
    确定所述采集到的声音中所述动力系统产生的噪声的特征参数；
    根据所述动力系统产生的噪声的特征参数确定所述补偿声音的特征参数。
  23. 根据权利要求22所述的无人机，其特征在于，所述处理器具体用于：
    对所述声音采集设备采集到的声音进行频域变换获取所述采集到的声音的频域数据；
    根据所述频域数据确定所述动力系统产生的噪声的特征参数。
  24. 根据权利要求23所述的无人机，其特征在于，所述处理器具体用于：
    从所述采集到的声音的频域数据中确定所述噪声对应的频域数据；
    根据所述噪声对应的频域数据确定所述动力系统产生的噪声的特征参数。
  25. 根据权利要求24所述的无人机，其特征在于，所述处理器具体用于：
    获取两个声音采集设备采集到的声音；
    对所述两个声音采集设备中每一个声音采集设备采集到的声音进行频域变换，获取采集到的两组声音的频域数据；
    根据所述两组声音的频谱数据和所述动力系统在所述无人机上的安装位置确定所述两组声音的频谱数据中所述噪声对应的频域数据分量。
  26. 根据权利要求25所述的无人机，其特征在于，所述声音采集设备为麦克风阵列。
  27. 根据权利要求18所述的无人机，其特征在于，所述补偿声音的频率与所述噪声的频率相同，且所述补偿声音的相位与所述噪声的相位相反。
  28. 根据权利要求18或27所述的无人机，其特征在于，所述补偿声音的频率与所述噪声的频率相同，且所述补偿声音的相位与所述噪声的相位相反，且所述补偿声音的振幅与所述噪声的振幅相同。
  29. 一种降噪装置，其特征在于，包括：存储器和处理器；
    所述存储器，用于存储程序代码；
    所述处理器，调用所述程序代码，当所述程序代码被执行时，用于执行以下操作：
    获取无人机在飞行过程中采集到的声音，其中，所述采集到的声音包括环境声源产生的声音和无人机在飞行过程中无人机的动力系统产生的噪声；
    将所述采集到的声音输入神经网络模型以获取降噪后的声音；其中，所述神经网络模型用于消除所述采集到的声音中所述动力系统产生的噪声。
  30. 根据权利要求29所述的装置，其特征在于，所述神经网络模型是以多个不同场景下采集到的环境声源产生的声音为输出、以所述多个不同场景下采集到的所述环境声源产生的声音和无人机在飞行过程中动力系统产生的噪声的混合声音为输入训练获取的。
  31. 根据权利要求30所述的装置，其特征在于，所述无人机在飞行过程中动力系统产生的噪声包括无人机在飞行过程中多个飞行状态对应的动力系统产生的噪声。
  32. 根据权利要求31所述的装置，其特征在于，所述多个飞行状态包括：加速飞行状态、减速飞行状态、悬停状态、转向状态、上升飞行状态和下降飞行状态中的至少两个。
PCT/CN2018/088673 2018-05-28 2018-05-28 降噪方法、装置和无人机 WO2019227279A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2018/088673 WO2019227279A1 (zh) 2018-05-28 2018-05-28 降噪方法、装置和无人机
CN201880031271.XA CN110709921A (zh) 2018-05-28 2018-05-28 降噪方法、装置和无人机

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/088673 WO2019227279A1 (zh) 2018-05-28 2018-05-28 降噪方法、装置和无人机

Publications (1)

Publication Number Publication Date
WO2019227279A1 true WO2019227279A1 (zh) 2019-12-05

Family

ID=68698543

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/088673 WO2019227279A1 (zh) 2018-05-28 2018-05-28 降噪方法、装置和无人机

Country Status (2)

Country Link
CN (1) CN110709921A (zh)
WO (1) WO2019227279A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021248364A1 (zh) * 2020-06-10 2021-12-16 深圳市大疆创新科技有限公司 无人飞行器录音方法、装置、芯片、无人飞行器、系统
CN111899711A (zh) * 2020-07-30 2020-11-06 长沙神弓信息科技有限公司 一种无人机传感器振动噪声抑制方法

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103971908A (zh) * 2014-05-06 2014-08-06 国家电网公司 一种变压器噪声抑制方法
CN105228054A (zh) * 2015-10-15 2016-01-06 深圳市大疆创新科技有限公司 飞行装置、拍摄装置及其录音降噪装置和方法
CN105261359A (zh) * 2015-12-01 2016-01-20 南京师范大学 手机麦克风的消噪系统和消噪方法
CN105899965A (zh) * 2014-08-29 2016-08-24 深圳市大疆创新科技有限公司 用于采集音频数据的无人飞行器
CN106527478A (zh) * 2016-11-24 2017-03-22 深圳市道通智能航空技术有限公司 无人机现场声音获取方法与有声视频实现方法及相关装置
CN206179506U (zh) * 2016-11-11 2017-05-17 马彦亭 一种无人机噪音消减装置
CN106782497A (zh) * 2016-11-30 2017-05-31 天津大学 一种基于便携式智能终端的智能语音降噪算法
CN107230472A (zh) * 2017-06-30 2017-10-03 邢优胜 一种直升机驾驶舱内噪声主动控制方法及系统
CN107452389A (zh) * 2017-07-20 2017-12-08 大象声科（深圳）科技有限公司 一种通用的单声道实时降噪方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2354858A1 (en) * 2001-08-08 2003-02-08 Dspfactory Ltd. Subband directional audio signal processing using an oversampled filterbank
CN105405439B (zh) * 2015-11-04 2019-07-05 科大讯飞股份有限公司 语音播放方法及装置

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103971908A (zh) * 2014-05-06 2014-08-06 国家电网公司 一种变压器噪声抑制方法
CN105899965A (zh) * 2014-08-29 2016-08-24 深圳市大疆创新科技有限公司 用于采集音频数据的无人飞行器
CN105228054A (zh) * 2015-10-15 2016-01-06 深圳市大疆创新科技有限公司 飞行装置、拍摄装置及其录音降噪装置和方法
CN105261359A (zh) * 2015-12-01 2016-01-20 南京师范大学 手机麦克风的消噪系统和消噪方法
CN206179506U (zh) * 2016-11-11 2017-05-17 马彦亭 一种无人机噪音消减装置
CN106527478A (zh) * 2016-11-24 2017-03-22 深圳市道通智能航空技术有限公司 无人机现场声音获取方法与有声视频实现方法及相关装置
CN106782497A (zh) * 2016-11-30 2017-05-31 天津大学 一种基于便携式智能终端的智能语音降噪算法
CN107230472A (zh) * 2017-06-30 2017-10-03 邢优胜 一种直升机驾驶舱内噪声主动控制方法及系统
CN107452389A (zh) * 2017-07-20 2017-12-08 大象声科（深圳）科技有限公司 一种通用的单声道实时降噪方法

Also Published As

Publication number Publication date
CN110709921A (zh) 2020-01-17

Similar Documents

Publication Publication Date Title
US11181809B2 (en) Focusing method, imaging device, and unmanned aerial vehicle
CN205263655U (zh) 一种用于自动生成全景照片的系统、无人飞行器及地面站
WO2019223270A1 (zh) 云台电机角度和角速度估算方法、装置、云台及飞行器
CN105045279A (zh) 一种利用无人飞行器航拍自动生成全景照片的系统及方法
JP7126143B2 (ja) 無人飛行体、情報処理方法およびプログラム
WO2020172800A1 (zh) 可移动平台的巡检控制方法和可移动平台
WO2018053825A1 (zh) 对焦方法和装置、图像拍摄方法和装置及摄像系统
WO2019227289A1 (zh) 延时拍摄控制方法和设备
WO2021217371A1 (zh) 可移动平台的控制方法和装置
WO2020062089A1 (zh) 磁传感器校准方法以及可移动平台
WO2020019212A1 (zh) 视频播放速度控制方法及系统、控制终端和可移动平台
WO2019104583A1 (zh) 最高温度点跟踪方法、装置和无人机
WO2020019260A1 (zh) 磁传感器校准方法、控制终端以及可移动平台
WO2019075758A1 (zh) 成像控制方法、成像装置和无人机
WO2020042159A1 (zh) 一种云台的转动控制方法、装置及控制设备、移动平台
WO2019227279A1 (zh) 降噪方法、装置和无人机
WO2019241970A1 (zh) 无人机的扬声器控制方法和设备
WO2019227287A1 (zh) 无人机的数据处理方法和设备
WO2019183789A1 (zh) 无人机的控制方法、装置和无人机
WO2021251441A1 (ja) 方法、システムおよびプログラム
US20210209133A1 (en) Data processing method and mobile platform
WO2020062024A1 (zh) 基于无人机的测距方法、装置及无人机
JP7501535B2 (ja) 情報処理装置、情報処理方法、情報処理プログラム
WO2020168519A1 (zh) 拍摄参数的调整方法、拍摄设备以及可移动平台
WO2020237429A1 (zh) 遥控设备的控制方法和遥控设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18920217

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18920217

Country of ref document: EP

Kind code of ref document: A1