Noises that are to be emitted by an aerial vehicle during operations may be predicted using one or more machine learning systems, algorithms or techniques. Anti-noises having equal or similar intensities and equal but out-of-phase frequencies may be identified and generated based on the predicted no
Noises that are to be emitted by an aerial vehicle during operations may be predicted using one or more machine learning systems, algorithms or techniques. Anti-noises having equal or similar intensities and equal but out-of-phase frequencies may be identified and generated based on the predicted noises, thereby reducing or eliminating the net effect of the noises. The machine learning systems, algorithms or techniques used to predict such noises may be trained using emitted sound pressure levels observed during prior operations of aerial vehicles, as well as environmental conditions, operational characteristics of the aerial vehicles or locations of the aerial vehicles during such prior operations. Anti-noises may be identified and generated based on an overall sound profile of the aerial vehicle, or on individual sounds emitted by the aerial vehicle by discrete sources.
대표청구항▼
1. An unmanned aerial vehicle (UAV) comprising: a frame;a plurality of motors mounted to the frame;a plurality of propellers, wherein each of the plurality of propellers is coupled to one of the plurality of motors;an audio speaker mounted to the frame; anda computing device having a memory and one
1. An unmanned aerial vehicle (UAV) comprising: a frame;a plurality of motors mounted to the frame;a plurality of propellers, wherein each of the plurality of propellers is coupled to one of the plurality of motors;an audio speaker mounted to the frame; anda computing device having a memory and one or more computer processors,wherein the one or more computer processors are configured to at least: determine a position of the UAV;determine an environmental condition associated with the position;determine an operating characteristic of at least one of the plurality of motors or at least one of the plurality of propellers associated with the position;identify a first sound pressure level and a first frequency of a first noise associated with the UAV based at least in part on at least one of the position, the environmental condition, or the operating characteristic;identify a second sound pressure level and a second frequency of a second noise previously captured using the microphone;determine a prior position of the UAV when the second noise was captured;determine a prior environmental condition associated with the prior position when the second noise was captured;determine a prior operating characteristic of the at least one of the plurality of motors or at least one of the plurality of propellers associated with the prior position when the second noise was captured;train a machine learning system based at least in part on information regarding the second sound pressure level, the second frequency, the prior position, the prior environmental condition and the prior operating characteristic;define a sound model for the UAV using the trained machine learning system; anddetermine a third sound pressure level of an anti-noise and a third frequency of the anti-noise corresponding to the first noise according to the sound model, wherein the third sound pressure level is not greater than the first sound pressure level, and wherein the third frequency approximates the first frequency and is substantially one hundred eighty degrees out of phase with the first frequency; andemit the anti-noise from the audio speaker of the UAV. 2. The UAV of claim 1, wherein the environmental condition at the position comprises at least one of: a temperature at the position;an atmospheric pressure at the position;a humidity at the position;a wind velocity at the position;a level of cloud cover at the position;a level of sunshine at the position; ora ground condition at the position,wherein the operating characteristic of at least one of the plurality of motors or at least one of the plurality of propellers at the position comprises at least one of:a course of the aerial vehicle at the position;an air speed of the aerial vehicle at the position;an altitude of the aerial vehicle at the position;a climb rate of the aerial vehicle at the position;a descent rate of the aerial vehicle at the position;a turn rate of the aerial vehicle at the position;an acceleration of the aerial vehicle at the position;a rotating speed of the at least one of the plurality of motors at the position; ora rotating speed of the at least one of the plurality of rotors at the position. 3. The UAV of claim 1, wherein the trained machine learning system is configured to perform at least one of: an artificial neural network;a conditional random field;a cosine similarity analysis;a factorization method;a K-means clustering analysis;a latent Dirichlet allocation;a latent semantic analysis;a log likelihood similarity analysis;a nearest neighbor analysis;a support vector machine; ora topic model analysis. 4. A method to operate a first aerial vehicle, the method comprising: identifying a first sound associated with at least one of a first position of the first aerial vehicle, a first operating characteristic of the first aerial vehicle at the first position, or a first environmental condition at the first position using at least one computer processor;providing information regarding the first sound to at least one machine learning system as an input, wherein the information regarding the first sound comprises at least one of a first sound pressure level of the first sound, a first frequency of the first sound, the first position, the first operating characteristic, or the first environmental condition;receiving, from the at least one machine learning system as an output, information regarding a second sound not more than twenty-five microseconds after the information regarding the first sound is provided to the at least one machine learning system as the input, wherein the information regarding the second sound comprises a second sound pressure level and a second frequency, and wherein the second frequency is substantially equal in magnitude and of reverse polarity with respect to the first frequency; andemitting the second sound with a first sound emitter of the first aerial vehicle. 5. The method of claim 4, wherein the at least one machine learning system is configured to perform at least one of: an artificial neural network;a conditional random field;a cosine similarity analysis;a factorization method;a K-means clustering analysis;a latent Dirichlet allocation;a latent semantic analysis;a log likelihood similarity analysis;a nearest neighbor analysis;a support vector machine; ora topic model analysis. 6. A method to operate a first aerial vehicle, the method comprising: providing information regarding a first sound to at least one machine learning system as a training input, wherein the information regarding the first sound comprises at least one of a first position associated with the first sound, a first operating characteristic associated with the first sound, or a first environmental condition associated with the first sound;providing information regarding a first sound pressure level of the first sound and a first frequency of the first sound to the at least one machine learning system as a training output;training the at least one machine learning system based at least in part on the training input and the training output;identifying a second sound associated with at least one of a second position of the first aerial vehicle, a second operating characteristic of the first aerial vehicle at the second position, or a second environmental condition at the second position;providing information regarding the second sound to the at least one trained machine learning system as an input, wherein the information regarding the second sound comprises at least one of a second sound pressure level of the second sound, a second frequency of the second sound, the second position, the second operating characteristic, or the second environmental condition;receiving, from the at least one trained machine learning system, information regarding a third sound as an output, wherein the information regarding the third sound comprises a third sound pressure level and a third frequency, and wherein the third frequency is substantially equal in magnitude and of reverse polarity with respect to the second frequency; andemitting the third sound with a first sound emitter of the first aerial vehicle. 7. The method of claim 6, wherein providing the information regarding the first sound to the at least one machine learning system as the training input comprises: determining the first position, wherein the first position is a position of the first aerial vehicle;determining the first operating characteristic associated with the first sound using the first aerial vehicle at the first position;determining the first environmental condition associated with the first sound using the first aerial vehicle at the first position; anddetermining the first sound pressure level of the first sound and the first frequency of the first sound using the first aerial vehicle at the first position. 8. The method of claim 6, wherein at least one of the first sound pressure level, the first frequency, the first operating characteristic, the first environmental condition, or the first position was determined at least in part by at least a second aerial vehicle. 9. The method of claim 6, wherein the at least one machine learning system is configured to perform at least one of: an artificial neural network;a conditional random field;a cosine similarity analysis;a factorization method;a K-means clustering analysis;a latent Dirichlet allocation;a latent semantic analysis;a log likelihood similarity analysis;a nearest neighbor analysis;a support vector machine; ora topic model analysis. 10. The method of claim 6, further comprising: identifying information regarding a transit plan for the first aerial vehicle, wherein the transit plan comprises information regarding a plurality of positions of the first aerial vehicle, and wherein the second position is one of the plurality of positions,wherein identifying the second sound associated with the at least one of the second position of the first aerial vehicle, the second operating characteristic of the first aerial vehicle at the second position, or the second environmental condition at the second position further comprises: identifying a plurality of sounds, wherein each of the plurality of sounds is associated with at least one of the plurality of positions for the first aerial vehicle; andproviding information regarding the plurality of sounds to the at least one trained machine learning system as the input,wherein receiving the information regarding the third sound as the output further comprises: receiving, from the at least one trained machine learning system, information regarding a plurality of anti-noises as the output, wherein the third sound is one of the plurality of anti-noises, and wherein each of the plurality of anti-noises corresponds to the at least one of the plurality of positions, andwherein emitting the third sound with the first sound emitter provided on the first aerial vehicle further comprises: emitting the plurality of anti-noises with the first sound emitter provided on the first aerial vehicle, wherein each of the plurality of anti-noises is emitted at the corresponding at least one of the plurality of positions. 11. A method to operate a first aerial vehicle, the method comprising: identifying a first sound associated with at least one of a first position of the first aerial vehicle, a first operating characteristic of the first aerial vehicle at the first position, or a first environmental condition at the first position using at least one computer processor, wherein the first sound has a first sound pressure level and a first frequency;determining a noise threshold within a vicinity of the first position; anddetermining a second sound based at least in part on the first sound and the first noise threshold using the at least one computer processor, wherein the second sound comprises a second sound pressure level and a second frequency, wherein the second frequency is equal in magnitude and of reverse polarity with respect to the first frequency, and wherein a sum of the first sound pressure level and the second sound pressure level is less than the noise threshold at a predetermined time; andemitting the second sound with a first sound emitter of the first aerial vehicle. 12. The method of claim 11, wherein the first sound emitter comprises one of an audio speaker, a piezoelectric sound emitter or a vibration source provided on the first aerial vehicle. 13. The method of claim 11, wherein the first aerial vehicle is projected to be located at the first position at a first time, and wherein emitting the second sound with the first sound emitter provided on the first aerial vehicle further comprises at least one of: emitting the second sound with the first sound emitter when the first aerial vehicle is at the first position; oremitting the second sound with the first sound emitter at the first time. 14. The method of claim 13, wherein the first environmental condition comprises at least one of a first temperature, a first barometric pressure, a first wind speed, a first humidity, a first level of cloud coverage, a first level of sunshine, or a first surface condition at the first position at the first time. 15. The method of claim 13, wherein the first operational characteristic comprises at least one of a first rotating speed of a first motor provided on the first aerial vehicle at the first time, a first altitude of the first aerial vehicle at the first time, a first course of the first aerial vehicle at the first time, a first airspeed of the first aerial vehicle at the first time, a first climb rate of the first aerial vehicle at the first time, a first descent rate of the first aerial vehicle at the first time, a first turn rate of the first aerial vehicle at the first time, or a first acceleration of the first aerial vehicle at the first time. 16. The method of claim 11, wherein determining the second sound based at least in part on the first sound and the noise threshold comprises: providing information regarding the first sound to a trained machine learning system as an input; andreceiving, from the trained machine learning system as an output, information regarding the second sound, andwherein the trained machine learning system is configured to perform at least one of:an artificial neural network;a conditional random field;a cosine similarity analysis;a factorization method;a K-means clustering analysis;a latent Dirichlet allocation;a latent semantic analysis;a log likelihood similarity analysis;a nearest neighbor analysis;a support vector machine; ora topic model analysis. 17. A method comprising: determining a position of an operating aerial vehicle;determining a noise threshold within a vicinity of the position;identifying information regarding a first noise associated with the operating aerial vehicle at the position, wherein the information regarding the first noise comprises a frequency of the first noise and an intensity of the first noise;identifying information regarding a second noise associated with the operating aerial vehicle at the position, wherein the information regarding the second noise comprises a frequency of the second noise and an intensity of the second noise;determining information regarding a first anti-noise based at least in part on the information regarding the first noise and the noise threshold, wherein the information regarding the first anti-noise comprises a frequency of the first anti-noise and an intensity of the first anti-noise, wherein the frequency of the first anti-noise is substantially equal to and out-of-phase with the frequency of the first noise and wherein a sum of the intensity of the first noise and the intensity of the first anti-noise is less than the noise threshold;determining information regarding a second anti-noise based at least in part on the information regarding the second noise and the noise threshold, wherein the information regarding the second anti-noise comprises a frequency of the second anti-noise and an intensity of the second anti-noise, wherein the frequency of the second anti-noise is substantially equal to and out-of-phase with the frequency of the second noise and wherein a sum of the intensity of the second noise and the intensity of the second anti-noise is less than the noise threshold;emitting the first anti-noise from a first noise emitting device associated with the operating aerial vehicle at the position; andemitting the second anti-noise from a second noise emitting device associated with the operating aerial vehicle at the position. 18. The method of claim 17, wherein the first noise is associated with a first component of the operating aerial vehicle, wherein the first component is one of at least one propeller, at least one motor, or at least a portion of an airframe of the operating aerial vehicle, wherein the first noise emitting device is associated with the first component,wherein the second noise is associated with a second component of the operating aerial vehicle, wherein the second component is another one of the at least one propeller, the at least one motor or at least the portion of an airframe of the operating aerial vehicle, andwherein the second noise emitting device is associated with the second component. 19. The method of claim 17, further comprising: identifying information regarding a third noise associated with the operating aerial vehicle at the position, wherein the information regarding the third noise comprises a frequency of the third noise and an intensity of the third noise;determining information regarding a third anti-noise based at least in part on the information regarding the third noise and the noise threshold, wherein the information regarding the third anti-noise comprises a frequency of the third anti-noise and an intensity of the third anti-noise, wherein the frequency of the third anti-noise is substantially equal to and out-of-phase with the frequency of the third noise and wherein a sum of the intensity of the third noise and the intensity of the third anti-noise is less than the noise threshold; andemitting the third anti-noise from a third noise emitting device associated with the operating aerial vehicle at the position. 20. The method of claim 19, wherein the first noise is associated with a first component of the operating aerial vehicle, wherein the first component is a first one of at least one propeller, at least one motor, or at least a portion of an airframe of the operating aerial vehicle, wherein the first noise emitting device is associated with the first component,wherein the second noise is associated with a second component of the operating aerial vehicle, wherein the second component is a second one of the at least one propeller, the at least one motor, or at least the portion of an airframe of the operating aerial vehicle,wherein the second noise emitting device is associated with the second component,wherein the third noise is associated with a third component of the operating aerial vehicle, wherein the third component is a third one of the at least one propeller, the at least one motor, or at least the portion of the airframe of the operating aerial vehicle, andwherein the third noise emitting device is associated with the third component. 21. The method of claim 17, further comprising: determining the information regarding the first anti-noise based at least in part on the information regarding the first noise, the noise threshold and the position; anddetermining the information regarding the second anti-noise based at least in part on the information regarding the second noise, the noise threshold and the position. 22. The method of claim 21, further comprising: determining at least one environmental condition within a vicinity of the position, wherein the at least one environmental condition comprises at least one of: a temperature within the vicinity of the position;an atmospheric pressure within the vicinity of the position;a humidity within the vicinity of the position;a wind velocity within the vicinity of the position;a level of cloud cover within the vicinity of the position;a level of sunshine within the vicinity of the position; ora ground condition within the vicinity of the position,wherein the information regarding the first anti-noise is determined based at least in part on the information regarding the first noise, the position, the noise threshold and the at least one environmental condition, andwherein the information regarding the second anti-noise is determined based at least in part on the information regarding the second noise, the position, the noise threshold and the at least one environmental condition. 23. The method of claim 21, further comprising: determining at least one operational characteristic of the aerial vehicle within a vicinity of the position, wherein the at least one operational characteristic comprises at least one of: an altitude of the aerial vehicle;a course of the aerial vehicle;an air speed of the aerial vehicle;a climb rate of the aerial vehicle;a descent rate of the aerial vehicle;a turn rate of the aerial vehicle;an acceleration of the aerial vehicle;a first rotating speed of the first propeller; ora second rotating speed of the first motor,wherein the information regarding the first anti-noise is determined based at least in part on the information regarding the first noise, the position, the noise threshold and the at least one operational characteristic; andwherein the information regarding the second anti-noise is determined based at least in part on the information regarding the second noise, the position, the noise threshold and the at least one operational characteristic. 24. The method of claim 17, wherein determining the information regarding the first anti-noise based at least in part on the information regarding the first noise and the noise threshold comprises: providing the information regarding the first noise to a trained machine learning system as a first input; andreceiving, from the trained machine learning system as a first output, the information regarding the first anti-noise,wherein determining the information regarding the second anti-noise based at least in part on the information regarding the second noise and the noise threshold comprises:providing the information regarding the second noise to the trained machine learning system as a second input; andreceiving, from the trained machine learning system as a second output, the information regarding the second anti-noise, andwherein the trained machine learning system is configured to perform at least one of:an artificial neural network;a conditional random field;a cosine similarity analysis;a factorization method;a K-means clustering analysis;a latent Dirichlet allocation;a latent semantic analysis;a log likelihood similarity analysis;a nearest neighbor analysis;a support vector machine; ora topic model analysis.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (5)
Bozich Daniel J. (San Diego CA) Wagenfeld Robert (Westport CT), Active gas turbine (jet) engine noise suppression.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.