JP4989553B2 - Vehicle noise reduction device and vehicle running noise reduction method - Google Patents

Description

  The present invention relates to a vehicle noise reduction device and a vehicle running noise reduction method.

  Conventionally, an active type that collects a noise signal, artificially generates a control signal having the same amplitude and opposite phase as the collected noise signal, and superimposes the control signal and the noise signal to control the noise signal. There is something.

  There is a noise reduction device that uses such a principle to reduce noise in a vehicle cabin (see, for example, Patent Documents 1 to 3).

  In Patent Document 1, engine noise is detected, a microphone is placed on the ceiling of the passenger compartment, etc., and residual noise is detected. The residual noise has zero amplitude and anti-phase so that the residual noise is zero. A drive signal is generated, and a control sound is output from the loudspeaker based on the generated drive signal.

  In the case of Patent Document 2, an acceleration sensor is installed on a passenger compartment panel to detect noise, a microphone and a speaker for canceling noise are installed in a seat, and the noise is set so that the interference sound detected by the microphone becomes zero. Is generated and supplied to the speaker.

In Patent Document 3, engine noise is detected, and a control sound is generated from a speaker so that a signal of a microphone arranged in a vehicle interior becomes zero.
Japanese Patent Laid-Open No. 5-46186 (page 4-6, FIGS. 1 and 9) Japanese Patent Laid-Open No. 6-110474 (page 4-6, FIG. 1) JP 2003-202871 A (page 5-9, FIGS. 1 and 9)

  However, noise during vehicle travel includes noise due to engine sound, wind noise, and tire travel noise.

  The sound insulation performance of the vehicle body has been improved year by year, and the noise caused by engine noise and wind noise has been considerably improved. However, tire running noise due to tire running sound has room for improvement compared to noise caused by engine noise and wind noise.

  The tire running noise is based on vibration generated when the tire tread pattern collides with the road surface when the vehicle is running. The vibration causes the surrounding air to vibrate while propagating through the vehicle body, resulting in tire running noise and reaching the driver's ear.

  Even considering only the vibration generated by one tire, it propagates to various locations in the body of the vehicle, so that driving noise can be heard from various directions as seen from the driver. In addition, due to the time for which the vibration propagates, the sounds emitted from various places on the body have different delay times. A combination of the sound from these various directions at the position of the driver's ear is the driving noise heard by the driver.

  If the tire location is different, the propagation situation also changes, so that different running noise occurs for each tire. The actual running noise is a combination of different running noises from all tires.

  Since the rotation speed of each tire is slightly deviated due to an inner ring difference or the like, the timing at which the tread pattern collides with the road surface changes for each tire. If the timing at which this vibration is generated is different, the vibration in which they are superposed is also different. Therefore, even if the vibration generated by each tire is substantially constant, the overlapped running noise changes.

  Due to the difference in the propagation time of vibration from each tire, the superimposed vibration also varies from place to place. In addition, when the vibration generation timing for each tire changes, how the superimposed vibration changes also varies from place to place.

  Consider the case where a microphone for noise collection is installed in the passenger compartment. The collected traveling noise and the traveling noise heard by the driver are different traveling noises due to different locations. Comparing components caused by vibration of one tire in running noise is complicated, but the propagation path is constant, so the correlation between the components does not change.

  However, when the vibration generation timing for each tire changes, the correlation between the superimposed traveling noises changes. Usually, since the tread pattern of each tire is the same, the vibration spectrum generated by each tire is the same, and the overlapped running noise cannot be decomposed for each tire.

  Since the correlation changes, it is impossible to learn the correlation in advance. Therefore, a device that actively cancels noise heard by the driver based on the sound collected by the microphone in the passenger compartment can be expected to be effective against engine noise having a single noise source. It cannot be effective against running noise that consists of noise sources.

  The present invention has been made in view of such conventional problems, and an object thereof is to provide a vehicle noise reduction device and a vehicle running noise reduction method capable of reducing tire running noise.

In order to achieve this object, a vehicle noise reduction device according to a first aspect of the present invention includes:
A plurality of microphones for collecting tire running noise, which individually collects the running noise generated by the tires of the vehicle and acquires a tire running noise signal;
Noise canceling sound signal generating signal processing means for generating a noise canceling sound signal based on a plurality of tire traveling noises individually picked up by the plurality of tire traveling noise sound collecting microphones, and
A noise canceling sound that outputs a noise canceling sound for canceling each tire running noise reaching the ears of the passengers in the vehicle based on the noise canceling sound signal generated by the signal processing means for generating the noise canceling sound signal An output speaker,
The signal processing means for generating the noise canceling sound signal is:
Correlation between tire traveling noise collected individually by the plurality of tire traveling noise collecting microphones and sound caused by tire traveling noise reaching the passenger's ear, and output from the noise canceling sound output speaker The noise canceling sound signal is generated based on the correlation between the noise canceling sound generated and the sound resulting from the noise canceling sound reaching the passenger's ear.

  The noise canceling sound output speakers may be individually provided to the left and right ears of the passenger in the vehicle.

The signal processing means for generating the noise canceling sound signal is:
A plurality of first filters representing a correlation between tire traveling noise individually picked up by the plurality of tire traveling noise sound collecting microphones and sound caused by tire traveling noise reaching the ear of the passenger;
A second filter representing a correlation between a noise cancellation sound output from the noise cancellation sound output speaker and a noise resulting from the noise cancellation sound reaching the passenger's ear,
The individual tire running noise signals collected by the tire running noise collecting microphone are passed through the first filters to generate individual tire running noise processing signals,
This individual tire running noise processing signal is added to generate an added signal after tire running noise processing,
The added signal after the tire running noise process may be passed through the second filter to generate the noise canceling sound signal.

Virtual tire running noise generating means for generating virtual tire running noise as a virtual noise source;
Tire vibration means for vibrating each tire based on the virtual tire running noise generated by the virtual tire running noise generating means;
An ear microphone that picks up sound reaching the passenger's ear by virtue of the tire vibration means vibrating each tire; and
The signal processing means for generating the noise canceling sound signal is:
First delay means for generating a tire running noise delay signal by delaying a sound signal by the tire vibration means picked up by the tire running noise collecting microphone;
A first filter coefficient calculation unit that sets a coefficient of the first filter based on a difference between a sound signal from the tire excitation unit picked up by the ear microphone and the tire running noise delay signal; You may prepare.

Virtual noise canceling sound signal generating means as a virtual sound source;
Based on the virtual noise cancellation sound signal, the noise cancellation sound output speaker that outputs a virtual noise cancellation sound;
An ear microphone that picks up the virtual noise canceling sound reaching the passenger's ear, and
The signal processing means for generating the noise canceling sound signal is:
Second delay means for delaying the virtual noise cancellation sound signal to generate a virtual noise cancellation sound delay signal;
A second filter coefficient calculation unit that sets a coefficient of the second filter based on a difference between the virtual noise cancellation sound delay signal and the virtual noise cancellation sound collected by the ear microphone. Also good.

  Each of the plurality of tire traveling noise collecting microphones may be attached in the vicinity of a holding portion of each tire that collects tire traveling noise.

  The noise canceling sound output speaker may be disposed in the vicinity of the passenger's ear.

  The tire vibration means may have a shape inserted between the tires and the ground.

The signal processing means for generating the noise canceling sound signal is:
An individual audio signal filter in which a coefficient is set based on the audio signal sound of each channel output from the audio speaker picked up by the individual microphone for picking up tire running noise;
An individual audio signal subtracting unit that subtracts an output signal of the individual audio signal filter from the individual tire running noise signal, and
You may make it use the output of the said separate audio signal subtraction part as an input of each said 1st filter.

Virtual audio signal generation means for generating a virtual audio signal as a virtual sound source,
The audio signal filter is:
A coefficient is determined based on a difference between the virtual audio signal generated by the virtual audio signal generation unit and the virtual audio signal acquired by the individual tire traveling noise pickup microphone and the virtual audio signal. You may make it do.

The signal processing means for generating the noise canceling sound signal is:
A plurality of specific frequency sound signal acquisition units for acquiring a sound signal of a specific frequency from each tire travel noise signal acquired by each of the tire travel noise sound pickup microphones;
A specific frequency sound signal subtracting unit that subtracts the specific frequency sound signal acquired by the specific frequency sound signal acquisition unit from each tire traveling noise signal acquired by each of the tire traveling noise sound collecting microphones,
You may make it use the output of the said specific frequency sound signal subtraction part as an input of each said 1st filter.

A traveling noise reduction method for a vehicle according to a second aspect of the present invention includes:
A plurality of microphones for collecting tire running noise, which individually collects the running noise generated by the tires of the vehicle and acquires a tire running noise signal;
Noise canceling sound for canceling the traveling noise of each tire reaching the ear of the passenger in the vehicle based on the noise canceling sound signal generated based on the traveling noise of the plurality of individually collected tires A noise canceling sound output speaker that outputs a noise cancellation method for a vehicle,
Correlation between individual tire vibration and sound caused by tire vibration reaching the passenger's ear, noise canceling sound output from the noise canceling sound output speaker, and noise canceling sound reaching the passenger's ear The noise canceling sound signal is generated based on the correlation with the sound caused by the noise.

The vehicle further includes:
A plurality of first filters representing a correlation between tire traveling noise individually picked up by the plurality of tire traveling noise sound collecting microphones and sound caused by tire traveling noise reaching the ear of the passenger;
A second filter representing a correlation between a noise canceling sound output from the noise canceling sound output speaker and a sound resulting from the noise canceling sound reaching the passenger's ear;
Virtual tire running noise generating means for generating virtual tire running noise as a virtual noise source;
Tire vibration means for vibrating each tire based on the virtual tire running noise generated by the virtual tire running noise generating means;
An ear microphone that picks up sound reaching the passenger's ear by virtue of the tire vibration means vibrating each tire;
Virtual noise canceling sound signal generating means as a virtual sound source;
Based on the virtual noise cancellation sound signal, the noise cancellation sound output speaker that outputs a virtual noise cancellation sound;
An ear microphone that picks up the virtual noise canceling sound reaching the passenger's ear, and
Delaying a sound signal by the tire vibration means picked up by the tire traveling noise sound collecting microphone to generate a tire traveling noise delay signal;
Setting a coefficient of the first filter based on a difference between a sound signal from the tire excitation means picked up by the ear microphone and the tire running noise delay signal;
Delaying the virtual noise cancellation sound signal to generate a virtual noise cancellation sound delay signal;
A step of setting a coefficient of the second filter based on a difference between the virtual noise cancellation sound delay signal and the virtual noise cancellation sound collected by the ear microphone.

  According to the present invention, tire running noise can be reduced.

Hereinafter, a vehicle noise reduction device according to an embodiment of the present invention will be described with reference to the drawings.
(Embodiment 1)
The configuration of the vehicle noise reduction device according to the first embodiment is shown in FIG.
The vehicle noise reduction device is configured to actively reduce the tire running noise of the vehicle 1, and includes sound collection microphones 11-1 to 11-4 and canceling sound generation speakers 12-1 and 12-2. And a signal processing unit 17.

  The signal processing unit 17 includes a plurality of filters that perform a filter process of the tire running noise signal for each tire based on the set filter coefficient. Then, as filter characteristics of each filter, filter characteristics (impulse response) from the sound pickup microphones 11-1 to 11-4 to the ears of the driver 2 as a passenger, and a speaker 12 for generating a cancellation sound at the driver's seat, respectively. The filter characteristics from −1, 12-2 to the ear of the driver 2 are set by learning.

  The vehicle noise reduction device further includes virtual traveling noise output vibrators 13-1 to 13-4, a virtual traveling noise generator 14, and a virtual canceling sound signal generator that are used when the signal processing unit 17 learns. 15, ear microphones 16-1 and 16-2, and a control unit 18.

  The signal processing unit 17 supplies the canceling sound signals Sc1 and Sc2 processed by the filter having the filter coefficient set based on the learned result to the canceling sound generating speakers 12-1 and 12-2. This vehicle noise reduction device is thus configured to reduce tire running noise.

  The sound collecting microphones 11-1 to 11-4 and the canceling sound generating speakers 12-1 and 12-2 shown in FIG. 1 are used in a normal operation state. The virtual running noise output vibrators 13-1 to 13-4, the virtual running noise generator 14, the virtual canceling sound signal generator 15, and the ear microphones 16-1 and 16-2 include the filter. Used for characteristic learning and removed in normal operating conditions.

  The sound collecting microphones 11-1 to 11-4 are microphones that individually collect the tire traveling noise of the tires 101 to 104 and acquire the tire traveling noise signals St1 to St4.

  Tire running noise is generated between the tires 101 to 104 and the road surface, and reaches the ears of the driver 2 through separate routes. Further, since the canceling sound generation speakers 12-1 and 12-2 are separated from the passenger's ear, it is necessary to predict the tire running noise that actually reaches the ear.

  When the tire running noise reaching the ear is predicted, since the delay time and the like are different for each route, it is necessary to process the tire running noise for each of the tires 101 to 104. However, when the tire running noises are mixed, it is difficult to decompose the mixed tire running noise for each of the tires 101 to 104.

  Therefore, it is desirable that the collection of the tire running noise is performed before mixing with the tire running noise from other tires.

  For this reason, the four sound pickup microphones 11-1 to 11-4 are provided, and the sound pickup microphones 11-1 to 11-4 have four tires 101 to 104, respectively. For example, in the vicinity of a shock absorber as a holding portion for each of the tires 101-104.

  The sound collecting microphones 11-1 to 11-4 supply the acquired tire running noise signals St 1 to St 4 to the signal processing unit 17.

  The canceling sound generating speakers 12-1 and 12-2 are based on the canceling sound signals Sc1 and Sc2 supplied from the signal processing unit 17, and the tire traveling noises for canceling the respective tire traveling noises that reach the ears of the driver 2. This is a speaker that outputs a canceling sound.

  The canceling sound generating speakers 12-1 and 12-2 are preferably installed as close to the ears of the driver 2 as possible in order to eliminate as much as possible the influence of the reflected sound caused by the tire traveling noise canceling sound in the vehicle. However, if it is provided right next to the ear of the driver 2, the merchantability of the vehicle 1 is greatly impaired.

  For this reason, the canceling sound generation speakers 12-1 and 12-2 are attached to the headrest 111 and individually disposed near the left and right ears of the driver 2 as shown in FIGS.

  As shown in FIG. 1, the canceling sound generating speakers 12-1 and 12-2 are provided with a cavity 112. Since the tire running noise contains a lot of bass components, the cavity 112 can output sufficient bass components from the canceling sound generating speakers 12-1 and 12-2.

  When the canceling sound generating speakers 12-1 and 12-2 are arranged as shown in FIGS. 1 and 2, the distance between the canceling sound generating speakers 12-1 and 12-2 and the driver 2 is the physique of the driver 2 or the like. Although there are variations in sheet settings, etc., it is approximately 15 cm.

  This distance corresponds to a quarter wavelength of a sound of about 500 Hz. In order to remove the tire running noise, those that output a canceling sound in a sound range of at least about 500 Hz are used for the canceling sound generating speakers 12-1 and 12-2.

  If the phase of the canceling sound from the canceling sound generating speakers 12-1 and 12-2 and the tire traveling noise are shifted by a quarter wavelength, the tire traveling noise is not canceled, and therefore the canceling sound generating speaker. The canceling sound from 12-1 and 12-2 needs to be output before the time when the tire traveling noise reaches the ear of the driver 2.

  The virtual traveling noise output vibrators 13-1 to 13-4 are supplied with virtual traveling noise and vibrate the tires 101 to 104 based on the supplied virtual traveling noise. The virtual traveling noise output vibrators 13-1 to 13-4 are preferably installed near the ground contact area with the road surface so that sound propagates in the same manner as when the vehicle 1 travels. It has the shape inserted between 101-104 and the ground.

  The four tires 101 to 104 are independent noise sources as described above and have different propagation characteristics. Further, there is no situation where vibration is applied to only one tire during traveling. The learning of the filter characteristics needs to be performed individually for each of the tires 101 to 104, and for this reason, four virtual running noise output vibrators 13-1 to 13-4 are provided.

  However, since learning is performed for each tire, it is possible to reduce the number of virtual traveling noise output vibrators to one by moving the virtual traveling noise output vibrator for each learning.

  The virtual travel noise generator 14 generates virtual travel noise Sd for driving the virtual travel noise output vibrators 13-1 to 13-4. The virtual running noise generator 14 generates, as the virtual running noise Sd, white noise having substantially the same intensity at all frequencies or pink noise having a frequency spectrum in which the frequency density is a reciprocal of the frequency. The virtual running noise generator 14 supplies the generated virtual running noise Sd to the signal processing unit 17.

  The virtual canceling sound signal generator 15 is for generating a learning canceling sound signal Sc0 when learning the filter characteristics. Similar to the virtual running noise generator 14, the virtual canceling sound signal generator 15 also generates white noise or pink noise as the canceling sound signal Sc0.

  The ear microphones 16-1 and 16-2 are microphones for capturing the ear noise Ne1 reaching the left ear and the ear noise Ne2 reaching the right ear of the driver 2 sitting in the driver seat in the vehicle 1, respectively. Used to learn the filter characteristics.

  The reason why the driver 2 is seated in the driver's seat and learns the filter characteristics is that the size of the head of the driver 2 is not so small that it can be ignored with respect to the wavelength of the noise. This is because it is desirable to capture noise while the driver 2 is sitting in the driver's seat.

  In order to capture the noise reaching the ear of the driver 2, the position of the ear microphones 16-1 and 16-2 should be close to the ear of the driver 2. However, since it is not preferable for learning if it is affected by the characteristics of the ear or ear canal, the ear microphones 16-1 and 16-2 are preferably arranged at a location about several centimeters away from the ear.

  Therefore, as shown in FIG. 2, the ear microphones 16-1 and 16-2 have microphones attached in place of the headphone type speakers and are arranged in the air near the ears of the driver 2. Is desirable.

  The signal processing unit 17 generates noise canceling sound signals Sc1 and Sc2 based on the tire running noise signals supplied from the sound pickup microphones 11-1 to 11-4 during normal use.

  Further, the signal processing unit 17 performs processing of each signal according to a learning step described later during learning.

  As shown in FIG. 3, the signal processing unit 17 during normal use includes amplifiers 21-1 to 21-4, A / D converters 22-1 to 22-4, and FIR (Finite Impulse Response). Filters 23-1 to 23-8, adders 24-1 and 24-2, FIR filters 25-1 and 25-2, D / A converters 26-1 and 26-2, and an amplifier 27-1. , 27-2.

  The amplifiers 21-1 to 21-4 amplify the tire running noise signals St1 to St4 supplied from the sound pickup microphones 11-1 to 11-4, respectively. The amplified tire running noise signals St1 to St4 are amplified. Supplied to A / D converters 22-1 to 22-4.

  The A / D converters 22-1 to 22-4 convert the analog tire running noise signals St1 to St4 supplied from the amplifiers 21-1 to 21-4 into digital tire running noise signals St1 to St4, respectively. Is.

  The A / D converter 22-1 supplies the converted tire running noise signal St1 to the FIR filters 23-1, 23-5, and the A / D converter 22-2 sends the converted tire running noise signal St2 to the FIR. This is supplied to the filters 23-2 and 23-6.

  The A / D converter 22-3 supplies the converted tire running noise signal St3 to the FIR filters 23-3 and 23-7, and the A / D converter 22-4 sends the converted tire running noise signal St4 to the FIR. This is supplied to the filters 23-4 and 23-8.

  The FIR filters 23-1 to 23-8 respectively filter the tire running noise signals St1 to St4 supplied from the A / D converters 22-1 to 22-4 based on the set filter coefficients. Is.

・・・・・・（１） The input / output characteristics of the FIR filter are expressed by the following equation (1), where the value of the input signal at sampling time n is x (n) and the value of the output signal is y (n).

(1)

  The FIR filters 23-1 to 23-8 perform a filter process on the tire running noise signals St1 to St4 according to the equation (1).

  The FIR filters 23-1 to 23-8 perform such filter processing, so that the tire traveling noises individually picked up by the sound pickup microphones 11-1 to 11-4 and the tires that reach the passenger's ears are collected. An individual tire running noise processing signal representing a correlation with a sound caused by running noise is output.

  The FIR filters 23-1 to 23-4 supply the output signal to the adder 24-1, and the FIR filters 23-5 to 23-8 supply the output signal to the adder 24-2.

  The adder 24-1 adds the values of the output signals respectively supplied from the FIR filters 23-1 to 23-4, and adds the added signal (addition after processing the tire running noise) to the FIR filter 25-1. Supply.

  The adder 24-2 adds the values of the output signals respectively supplied from the FIR filters 23-5 to 23-8, and adds the added signal (addition after processing the tire running noise) to the FIR filter 25-2. Supply.

  The FIR filters 25-1 and 25-2 perform filter processing according to the equation (1) on the signals added by the adders 24-1 and 24-2, respectively. By performing such filter processing, the FIR filters 25-1 and 25-2 cause the tire traveling noise canceling sound output from the noise canceling sound generating speakers 12-1 and 12-2 and the driver 2 A (digital) signal representing a correlation with the sound caused by the noise canceling sound reaching the ear is supplied to the D / A converters 26-1 and 26-2.

  The D / A converters 26-1 and 26-2 convert the digital signals supplied from the FIR filters 25-1 and 25-2 into analog signals, respectively, and the converted analog signals are amplifiers 27-1. , 27-2.

  The amplifiers 27-1 and 27-2 amplify the analog signals supplied from the D / A converters 25-1 and 25-2, respectively.

  The signal processing unit 17 supplies the signals amplified by the amplifiers 27-1 and 27-2 to the canceling sound generating speakers 12-1 and 12-2 as canceling sound signals Sc1 and Sc2, respectively.

The signal processing unit 17 is controlled by the control unit 18 and executes learning steps 1 and 2 using the ear microphones 16-1 and 16-2, thereby allowing FIR filters 23-1 to 23-8 and 25-1. 25-2, the optimum filter coefficient H _j (n) is determined.

  As shown in FIG. 4, the learning step 1 supplies virtual running noise to the virtual running noise output vibrators 13-1 to 13-4 for each tire, and each sound collecting microphone 11-i (i = 1 to 1). 4), the filter characteristics from the sound pickup microphone 11-i to the ear microphone 16-1 on the left side of the driver 2 and the right side of the driver 2 from the sound pickup microphone 11-i based on the learning tire running noise signal acquired in 4). This is a step of individually learning the filter characteristics up to the ear microphone 16-2.

  In the learning step 2, the noise canceling sound signal Sc0 for learning is supplied to the noise canceling sound generating speakers 12-1 and 12-2 for noise canceling, and the driver seat canceling sound generating speakers 12-1 and 12- are supplied. 2 to individually learn the filter characteristics from the ear microphones 16-1 and 16-2.

  When performing the learning step 1 from the sound pickup microphone 11-i to the ear microphones 16-1 and (16-2) on the left side of the driver 2, the signal processing unit 17 performs D as shown in FIG. / A converter 31, amplifier 32, amplifier 21-i (i = 1 to 4), A / D converter 22-i, delay device 41, FIR filter 23-i, amplifier 42, An A / D converter 43, an adder 44, and an LMS (Least Mean Square) calculator 45 are included.

  The D / A converter 31 converts the digital virtual running noise Sd supplied from the virtual running noise generator 14 into an analog virtual running noise Sd, and supplies the converted virtual running noise Sd to the amplifier 32. .

  The amplifier 32 amplifies the virtual running noise Sd supplied from the D / A converter 31. The signal processing unit 17 supplies the virtual running noise Sd amplified by the amplifier 32 to the virtual running noise output shaker 13-i as the virtual running noise Sd1.

  The delay device 41 is supplied with the analog tire running noise signal Sti from the A / D converter 22-i and delays the supplied tire running noise signal Sti. This delayed signal corresponds to a tire running noise delay signal.

  This delay device 41 is provided for correcting the input timing to the FIR filter 23-i, and the delay time is the same as the delay device 51 of FIG.

  This delay time corresponds to the time from when the canceling sound is generated until the canceling sound output from the canceling sound generating speaker 12-i reaches the ear of the driver 2 and further reaches the FIR filter. The signal passing through the FIR filter 23-i whose coefficient is determined in the learning step 1 is finally used for generating a canceling sound.

  Since it takes time from the generation of the canceling sound until the noise is actually canceled at the ear, it is necessary to move forward by that time in the learning step. Therefore, in order to execute the learning step 1, this delay device 41 is necessary.

  The amplifier 42 is supplied with the ear noise Ne1 from the ear microphones 16-1 and (16-2) as an analog signal, and amplifies the ear noise Ne1. The amplified ear noise Ne1 is supplied to the A / D converter 43. Supply.

  The A / D converter 43 converts the analog ear noise Ne1 supplied from the amplifier 42 into digital ear noise Ne1, and supplies the converted ear noise Ne1 to the adder 44.

  The adder 44 and the LMS calculator 45 use the tire running noise signal Sti delayed by the delay unit 41 as a target signal, and the ear noise Ne1, (Ne2) supplied from the ear microphones 16-1, (16-2). Is used as a reference signal, and filter coefficients of FIR 23-1 to FIR 23-8 are set based on an error signal between the reference signal and the output signals of the filters 23-1 to 23-8.

  The adder 44 adds the values (+) of the ear noises Ne1 and Ne2 supplied from the A / D converter 43 and the value (-) of the signal output from the FIR filter 23-i. The added signal is used as an error signal, and its value e (n) is supplied to the LMS calculator 45.

The LMS calculator 45 updates the filter coefficient H _j (n) of the FIR filter 23-i according to the least square method algorithm.

・・・・・・（２）
ＬＭＳ演算器４５は、この数式（２）に従い、逐次、誤差信号の値ｅ（ｎ）のエネルギーが最小となるようなフィルタ係数Ｈ_j（ｎ）を求める。そして、ＬＭＳ演算器４５は、求めたフィルタ係数Ｈ_j（ｎ）をＦＩＲフィルタ２３−ｉに供給する。 The coefficient update formula of the filter coefficient H _j (n) of the FIR filter 23-i by the LMS algorithm is expressed by the following formula (2).

(2)
The LMS calculator 45 sequentially obtains a filter coefficient H _j (n) that minimizes the energy of the error signal value e (n) according to the equation (2). Then, the LMS calculator 45 supplies the obtained filter coefficient H _j (n) to the FIR filter 23-i.

The signal processing unit 17 configured as described above determines the optimum filter coefficient H _j (n) of the FIR filter 23-i by executing the learning step 1.

  Next, when executing the learning step 2 from the cancellation sound generating speakers 12-1 and (12-2) to the ear microphones 16-1 and (16-2), the signal processing unit 17 is configured as shown in FIG. , (FIG. 16B), D / A converters 26-1, (26-2), amplifiers 27-1, (27-2), a delay unit 51, an amplifier 52, An A / D converter 53, an adder 54, an LMS calculator 55, and FIR filters 25-1 and (25-2) are included.

  The delay unit 51 delays the canceling sound signal Sc0 for learning output from the virtual canceling sound signal generator 15 in order to correct the input timing to the FIR filters 25-1 and (25-2). The signal delayed by the delay unit 51 corresponds to a virtual noise canceling sound delay signal.

  The delay time of the sound from the virtual canceling sound signal generator 15 that generates a signal to be output to the speaker 12-1 for canceling sound at the driver's seat and (12-2) to the ear microphones 16-1 and (16-2) is , The processing time T1 of the D / A converters 26-1 and (26-2) and the sound from the cancellation sound generating speakers 12-1 and 12-2 to the ear microphones 16-1 and 16-2. It is a time (T1 + T2) obtained by adding the propagation time T2 to propagate.

The A / D converter 53 between the ear microphones 16-1 and (16-2) and the FIR filter 25-i also has a processing time T3, and the FIR filter 25 is used as a target signal from the virtual canceling sound signal generator 15. The delay time ΔT until it is input to −i is a time obtained by adding them (T1 + T2 + T3).
The delay time of the delay unit 51 is a length corresponding to ΔT.

  The amplifier 52, the A / D converter 53, the adder 54, and the LMS calculator 55 are respectively an amplifier 42, an A / D converter 43, an adder 44, and an LMS calculator shown in FIGS. 5 (a) and 5 (b). 45.

  The adder 54 and the LMS calculator 55 use the ear noises Ne1 and (Ne2) collected by the ear microphones 16-1 and 16-2 as target signals, and the noise canceling sound signal Sc0 delayed by the delay unit 51, respectively. As the reference signal, the filter coefficients of the filters 25-1 and (25-2) are set based on the error signal between the reference signal and the output signals of the filters 25-1 and (25-2).

  The adder 54 adds the value (+) of the noise canceling sound signal Sc0 delayed by the delay unit 51 and the value (-) of the output signal of the filters 25-1 and (25-2). The signal e (n) is supplied to the LMS calculator 55 as an error signal.

The LMS calculator 55 sequentially obtains a filter coefficient H _j (n) that minimizes the energy of the error signal value e (n) in accordance with Equation (2). Then, the LMS calculator 55 supplies the obtained filter coefficient H _j (n) to the FIR filter 25-1.

The signal processing unit 17 configured as above determines the optimum filter coefficient H _j (n) of the FIR filters 25-1 and (25-2) by executing the learning step 2.

  The control unit 18 controls the entire vehicle noise reduction device. The control unit 18 controls the virtual running noise generator 14, the virtual cancellation sound signal generator 15, and the signal processing unit 17 so that the signal processing unit 17 executes the learning steps 1 and 2.

  In addition, the control unit 18 controls the signal processing unit 17 so as to reduce tire running noise in the normal operation state.

Next, the operation of the vehicle noise reduction device according to the first embodiment will be described.
The operation of the vehicle noise reduction device is divided into normal use and learning. In addition, the learning includes two steps of learning step 1 for learning the characteristic from the sound collecting microphone to the ear of the driver 2 and learning step 2 for learning the characteristic from the canceling sound generating speaker 12-i to the ear of the driver 2. Divided into learning steps.

First, the operation of learning step 1 will be described.
As shown in FIG. 5A, the control unit 18 first uses the ear microphone 16-1 and the sound collection microphone 11-1 on the left side of the driver 2 to configure the FIR filter 23-1 of the signal processing unit 17. The filter characteristic is learned (i = 1 in FIG. 5A). This learning step 1 is defined as step S11.

  The control part 18 performs this step S11 as follows. First, the control unit 18 controls the virtual running noise generator 14 to generate digital virtual running noise Sd as a virtual noise source, and supplies the generated virtual running noise Sd to the signal processing unit 17.

  The D / A converter 31 of the signal processing unit 17 converts the digital virtual running noise Sd as the virtual noise source supplied from the virtual running noise generator 14 into an analog virtual running noise Sd.

  The amplifier 32 amplifies the virtual running noise Sd as the virtual noise source, and supplies the amplified virtual running noise Sd to the virtual running noise output shaker 13-1 disposed in the vicinity of the left front tire 101. . The virtual running noise output shaker 13-1 outputs a virtual noise sound from the tire 101 based on the supplied virtual running noise Sd.

  The sound collecting microphone 11-1 detects the virtual noise sound and supplies the virtual noise sound signal St1 to the amplifier 21-1 of the signal processing unit 17.

  The amplifier 21-1 amplifies the virtual noise sound signal St1, and the A / D converter 22-1 converts the virtual noise sound signal St1 into a digital signal.

  The delay device 41 delays the virtual noise sound signal St1, and supplies the delayed virtual noise sound signal St1 to the FIR filter 23-1. The FIR filter 23-1 uses the virtual noise sound signal St1 as an input signal, and outputs an output signal whose value is obtained according to Equation (1).

  On the other hand, the amplifier 42 amplifies the ear noise Ne1 supplied from the ear microphone 16-1, and the A / D converter 43 converts the ear noise Ne1 into the digital ear noise Ne1.

  The adder 44 adds this value and the value y (n) of the output signal of the FIR filter 23-1 using the ear noise Ne1 as a reference signal, and supplies the added signal to the LMS computing unit 45 as an error signal. To do.

The LMS calculator 45 sequentially updates the filter coefficient H _i (n) of the FIR filter 23-1 based on the error signal supplied from the adder 44 in accordance with Expression (2).

The control unit 18 fixes the filter coefficient H _j (n) of the FIR filter 23-1 when the error signal value e (n) from the adder 44 of the signal processing unit 17 is minimized. Thus, the control unit 18 determines the filter coefficient H _j (n) of the FIR filter 23-1 of the signal processing unit 17 by executing Step S11.

  Similarly, the left and right ear microphones 16-1 and 16-4 for the virtual traveling noise output vibrators 13-1 to 13-4 and the sound collecting microphones 11-1 to 11-4 corresponding to the four tires, front, rear, left, and right, respectively. The learning step 1 of 8 routes from 16 to 16-2 is executed as steps S11 to S18.

  Since a plurality of the virtual running noise output vibrators 13-i are not used at the same time, the D / A converter 31, the amplifier 32, and the virtual running noise output vibrator 13 are only one set, Each time the route to be learned is changed, the location of the virtual running noise output vibrator 13 may be changed and advanced.

  Second, the operation of learning step 2 will be described. As shown in FIG. 6A, the control unit 18 uses the ear microphone 16-i of the driver 2 to cause the signal processing unit 17 to filter from the cancellation sound generating speaker 12-i to the ear microphone 16-i. Learn characteristics. This learning step 2 is defined as step S21.

The control part 18 performs this step S21 as follows.
First, the control unit 18 controls the virtual canceling sound signal generator 15 to generate the virtual canceling sound signal Sc0, and supplies the generated virtual canceling sound signal Sc0 to the signal processing unit 17.

  The D / A converter 26-1 of the signal processing unit 17 converts the digital virtual canceling sound signal Sc0 supplied from the virtual canceling sound signal generator 15 into an analog virtual canceling sound signal Sc0.

  The amplifier 27-1 amplifies the virtual canceling sound signal Sc0 and supplies the amplified virtual canceling signal Sc1 to the canceling sound generating speaker 12-1.

  The canceling sound generating speaker 12-1 outputs a virtual canceling sound based on the supplied virtual canceling signal Sc1, and the ear microphone 16-1 outputs an analog virtual canceling sound output from the canceling sound generating speaker 12-1. Enter. Then, the ear microphone 16-1 supplies a virtual cancellation signal corresponding to the ear noise Ne <b> 1 to the amplifier 52.

  The amplifier 52 amplifies the ear noise Ne1, and the A / D converter 53 converts the analog ear noise Ne1 into a digital ear noise Ne1, and converts the converted digital ear noise Ne1 into the FIR filter 25-1. Supply.

  The FIR filter 25-1 receives the ear noise Ne1 as an input signal, and outputs an output signal whose value y (n) is obtained in accordance with Equation (1).

  On the other hand, the delay unit 51 delays the virtual canceling sound signal Sc0 supplied from the virtual canceling sound signal generator 15 by a predetermined time and supplies the delayed signal to the adder 54.

  The adder 54 adds the value y (n) of the output signal of the FIR filter 25-1 and the virtual canceling sound signal Sc0 supplied from the delay unit 51, and uses the added value as an error signal as an error signal. To supply.

The LMS calculator 55 updates the filter coefficient H _j (n) of the FIR filter 25-1 based on the error signal value e (n) according to the equation (2).

The control unit 18 fixes the filter coefficient H _j (n) of the FIR filter 25-1 when the error signal value e (n) from the adder 54 of the signal processing unit 17 is minimized. In this way, the control unit 18 determines the filter coefficient H _j (n) of the FIR filter 25-1 of the signal processing unit 17 by executing step S21.

  Similarly, the control unit 18 learns the filter characteristics from the canceling sound generating speaker 12-2 to the ear microphone 16-2 as step S22, and calculates the filter coefficient Hj (n) of the FIR filter 25-2. decide.

Third, the operation in the normal use state will be described.
Based on the optimum filter coefficients H _j (n) of the FIR filters 23-1 to 23-8, 25-1, 25-2, the signal processing unit 17 has the same phase and opposite phase as the tire running noise at the driver's ear The canceling sound signals Sc1 and Sc2 having the amplitude are generated. The signal processing unit 17 supplies the generated canceling sound signals Sc1 and Sc2 to the canceling sound generating speakers 12-1 and 12-2, respectively.

  The canceling sound generating speakers 12-1 and 12-2 output canceling sounds based on the supplied canceling sound signals Sc1 and Sc2, respectively. Thereby, the tire running noise is canceled and reduced at the driver's ear.

  As described above, according to the first embodiment, the sound collection microphones 11-1 to 11-4 individually collect the tire running noise of the tires 101 to 104 and reduce the tire running noise. .

  Accordingly, it is possible to effectively reduce a plurality of tire running noises in which the tires 101 to 104 are independent.

Further, the signal processing unit 17 executes learning step 1 for each of the tires 101 to 104 using the ear microphones 16-1 and 16-2, and FIR filters 23-1 to 23-8, 25-1, and 25-2. Filter coefficient H _j (n).

Further, the signal processing unit 17 executes the learning step 2 for each of the cancellation sound generating speakers 12-1 and 12-2 using the ear microphones 16-1 and 16-2, and performs FIR filters 23-1 to 23-8. 25-1 and 25-2, the filter coefficients H _j (n) are determined.

Therefore, even if the tires 101 to 104 are a plurality of independent noise sources and each frequency band is the same, the optimum filter coefficient H _j (n) can be determined.

  In learning steps 1 and 2, delay devices 41 and 51 are provided, respectively, and the virtual noise sound signals St1 to St4 and the canceling sound signal Sc0 are delayed so as to delay the FIR filters 23-1 to 23-8, 25-1, and 25. The input timing to -2 was corrected.

  For this reason, it is possible to accurately match the timing of signal processing with respect to the travel time of travel noise, and to effectively reduce tire travel noise.

(Embodiment 2)
The vehicle noise reduction device according to the second embodiment eliminates the influence of the audio signal of the audio device, and reduces the tire running noise without canceling the audio signal.

  As shown in FIG. 7, the vehicle 1 according to the second embodiment includes an audio device 19. The audio device 19 includes speakers 60-1 and 60-2 and amplifiers 61-1 and 61-2.

  The speakers 60-1 and 60-2 are the L channel and R channel speakers of the audio device 19, respectively, and output the L channel and R channel sounds based on the signals supplied from the amplifiers 61-1 and 61-2. Output.

  The amplifier 61-1 amplifies the left audio signal S_ad (L) (analog signal) by (amplitude), and supplies the amplified audio signal S_ad (L) to the speaker 60-1.

  The amplifier 61-2 amplifies (amplifies) the right audio signal S_ad (R) (analog signal), and supplies the amplified audio signal S_ad (R) to the speaker 60-2.

  The signal processing unit 17 includes A / D converters 62-1 to 62-8, FIR filters 63-1 to 63-8 (indicated as FIR (Ax) in the figure), and adders 64-1 to 64-64. (The A / D converters 63-3 to 62-6 and the FIR filters 63-3 to 63-6 are not shown).

  The A / D converters 62-1, 62-3, 62-5, and 62-7 convert the analog audio signal S_ad (L) into the digital audio signal S_ad (L), and the converted audio signal. S_ad (L) is supplied to the FIR filters 63-1, 63-3, 63-5, and 63-7, respectively.

  The A / D converters 62-2, 62-4, 62-6, and 62-8 convert the analog audio signal S_ad (R) into a digital audio signal S_ad (R), and the converted audio signal. S_ad (R) is supplied to the FIR filters 63-2, 63-4, 63-6, and 63-8, respectively.

  The FIR filters 63-1 to 63-8 have the audio signals S_ad (L) and S_ad (R) supplied from the A / D converters 62-1 to 62-8 as input signals, respectively, according to the equation (1). And an audio signal filter for performing filter processing.

The adders 64-1 to 64-4 are output from the FIR filters 63-1 to 63-4, the signals output from 63-5 to 63-8, and the A / D converters 22-1 to 22-4. Are added to the FIR filters 23-1 to 23-4 and 23-5 to 23-8.
The outputs of the adders 64-1 to 64-4 are used as inputs to 23-1 to 23-4 and 23-5 to 23-8.

The signal processing unit 17 learns the filter characteristics from the audio signals S_ad (L) and S_ad (L) to the sound pickup microphones 11-1 to 11-4, respectively, and thereby the FIR filters 63-1 to 63-8. An optimum filter coefficient H _j (n) is determined.

  The vehicle 1 includes an audio signal generator 65 as shown in FIG. The audio signal generator 65 generates a learning audio signal S_ad0 as a virtual audio signal.

  The amplifier 61-p of the audio device 19 is supplied with the audio signal S_ad0 from the audio signal generator 65, amplifies the supplied audio signal S_ad0, and supplies the amplified audio signal S_ad0 to the speaker 60-p.

  As shown in FIG. 8, the signal processing unit 17 at the time of learning the filter characteristics further includes an adder 66 and an LMS calculator 67.

  The adder 66 adds the signal (+) supplied from the A / D converter 22-i and the signal (−) output from the FIR filter 63-k, and uses the added signal as an error signal. The value e (n) is supplied to the LMS calculator 67.

The LMS calculator 67 sequentially updates the filter coefficient H _j (n) of the FIR filter 63-k based on the equation (2) based on the error signal value e (n) supplied from the adder 66. When the error signal value e (n) is minimized, the filter coefficient H _j (n) of the FIR filter 63-k is fixed.

  The FIR filter 63-k corresponds to a filter that directly calculates a signal component from which the audio signal S_ad0 output from the speaker 60-p is input through the sound pickup microphone 11-i from the audio signal S_ad0. Therefore, the audio signal S_ad0 can be removed by processing the audio signal S_ad0 with the FIR filter 63-k and subtracting it from the signal input through the sound pickup microphone 11-i.

In this way, the audio signal S_ad0 generated by the audio signal generator 65 is picked up by the individual pickup microphones 11-1 to 11-4 as a virtual audio pickup signal, and the FIR filters 63-1 to 63-63. The filter coefficient H _j (n) of −8 is determined based on the difference between the tire running noise signals St1 to St4 collected as the virtual audio pickup signal and the audio signal S_ad0.

When the signal processing unit 17 performs such learning and the filter coefficients H _j (n) of the FIR filters 63-1 to 63-8 are determined, the control unit 18 controls the signal processing unit 17 to perform a normal operation. Make it work.

  As described above, the adder 64-i in FIG. 7 performs an operation of subtracting the audio signal component calculated by the FIR filter 63-k from the signal collected by the sound collecting microphone 11-i. Thereby, it is possible to prevent the audio signal S_ad0 from being mixed into the cancellation sound signal generated thereafter. That is, when the tire running noise is canceled, it is possible to prevent the sound from the audio device 19 from being canceled at the same time.

  The flow after the FIR filters 23-1 to 23-4 and 23-5 to 23-8 generates and outputs canceling sound signals Sc 1 and Sc 2 as in the first embodiment. Thereby, tire running noise is reduced.

As described above, according to the second embodiment, the signal processing unit 17 learns the filter characteristics from the audio signals S_ad (L) and S_ad (L) to the sound pickup microphones 11-1 to 11-4. Thus, the optimum filter coefficient H _j (n) of the FIR filters 63-1 to 63-8 is determined.

  Therefore, it is possible to reduce the tire running noise by eliminating the influence of the audio signals S_ad (L) and S_ad (L) mixed in the tire running noise.

(Embodiment 3)
The vehicle noise reduction device according to the third embodiment is configured such that an alarm sound of an emergency vehicle mixed in tire running noise is not excluded as tire running noise.

  As illustrated in FIG. 9, the vehicle noise reduction device according to the third embodiment includes signal acquisition units 71-1 to 71-4 and adders 72-1 to 72-4.

  Each of the signal acquisition units 71-1 to 71-4 acquires a continuous signal having a specific frequency such as an alarm sound of an emergency vehicle from the signals output from the A / D converters 22-1 to 22-4, respectively. The digital filter allows only a continuous sound signal to pass at this specific frequency.

  The signal acquisition units 71-1 to 71-4 supply the acquired continuous signals to the adders 72-1 to 72-4, respectively.

  The adders 72-1 to 72-4 invert the signals supplied from the signal acquisition units 71-1 to 71-4, respectively, and the inverted signals and the A / D converters 22-1 to 22-4 output them. Are added together.

  For example, when an emergency vehicle passes nearby while outputting an alarm sound of a specific frequency, a continuous signal due to the alarm sound is mixed into tire running noise.

  The signal acquisition units 71-1 to 71-4 extract the continuous signals from the signals output from the A / D converters 22-1 to 22-4, respectively.

  The adders 72-1 to 72-4 invert the continuous signals acquired by the signal acquisition units 71-1 to 71-4, respectively, and convert the inverted signals to the A / D converters 22-1 to 22-1, respectively. It adds with the signal output from 22-4.

  By adding, the warning sound of the emergency vehicle is removed from the signal that generates the tire running noise canceling sound, and the adders 72-1 to 72-4 correspond to the specific frequency sound signal subtracting unit. The outputs of the adders 72-1 to 72-4 are used as inputs of the FIRs 23-1 to 23-8.

  For this reason, it is possible to avoid the warning sound of the emergency vehicle being canceled by the canceling sound. The processes after the FIR filters 23-1 to 23-8 are the same as those in the first embodiment.

  As described above, according to the third embodiment, the signal processing unit 17 acquires a continuous signal having a specific frequency from the signals output from the A / D converters 22-1 to 22-4, and inverts them to obtain A The signals output from the / D converters 22-1 to 22-4 are added.

  Therefore, the tire running noise can be reduced without being affected by the continuous signal mixed in the tire running noise.

In carrying out the present invention, various forms are conceivable and the present invention is not limited to the above embodiment.
In the above embodiment, the passenger is described as the driver 2. However, the passenger is not limited to the driver 2 and may be a passenger sitting in the passenger seat and the rear seat.

It is a block diagram which shows the structure of the vehicle noise reduction apparatus which concerns on Embodiment 1 of this invention. It is a figure which shows the headrest arrange | positioned around a driver, the speaker for cancellation noise generation, and the ear microphone. It is a figure which shows the structure of the signal processing part in a normal operation state. It is a figure which shows the learning step which the signal processing part shown in FIG. 1 performs. It is a figure which shows the structure of the signal processing part at the time of performing the learning step 1 from each sound collection microphone to each ear microphone, (a), (b) is the left side and right side of a driver from each sound collection microphone, respectively. It is a figure which shows the structure of the signal processing part in the case of performing the learning step 1 to the ear microphone. It is a figure which shows the structure of the signal processing part in the case of performing the learning step 2 which learns the filter characteristic from each speaker for cancellation noise generation of a driver's seat to each ear microphone, (a), (b), It is a figure which shows the structure of the signal processing part in the case of performing the learning step 2 which learns the filter characteristic from the left and right cancellation noise generating speakers to the left and right ear microphones of the driver's seat. It is a block diagram which shows the structure of the vehicle noise reduction apparatus which concerns on Embodiment 2 of this invention. It is a block diagram which shows the structure of the signal processing part in the case of performing a learning step. It is a block diagram which shows the structure of the noise reduction apparatus for vehicles which concerns on Embodiment 3 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Vehicle 2 Driver 17 Signal processing part 11-1 to 11-4 Sound collection microphones 12-1, 12-2 Speaker for canceling sound generation 13-1 to 13-4 Vibrator for virtual running noise output 14 Virtual running noise generation 15 Virtual canceling sound signal generator 16-1, 16-2 Ear microphone 23-1 to 23-8 FIR filter 25-1, 25-2 FIR filter 63-1 to 63-8 FIR filter 65 Audio signal generator

Claims (13)

A plurality of microphones for collecting tire running noise, which individually collects the running noise generated by the tires of the vehicle and acquires a tire running noise signal;
Noise canceling sound signal generating signal processing means for generating a noise canceling sound signal based on a plurality of tire traveling noises individually picked up by the plurality of tire traveling noise sound collecting microphones, and
A noise canceling sound that outputs a noise canceling sound for canceling each tire running noise reaching the ears of the passengers in the vehicle based on the noise canceling sound signal generated by the signal processing means for generating the noise canceling sound signal An output speaker,
The signal processing means for generating the noise canceling sound signal is:
Correlation between tire traveling noise collected individually by the plurality of tire traveling noise collecting microphones and sound caused by tire traveling noise reaching the passenger's ear, and output from the noise canceling sound output speaker Generating a noise canceling sound signal based on a correlation between the noise canceling sound to be generated and a sound resulting from the noise canceling sound reaching the passenger's ear,
A vehicle noise reduction device characterized by the above. The noise canceling sound output speaker is individually provided in each of the left and right ears of the passenger in the vehicle,
The vehicle noise reduction device according to claim 1. The signal processing means for generating the noise canceling sound signal is:
A plurality of first filters representing a correlation between tire traveling noise individually picked up by the plurality of tire traveling noise sound collecting microphones and sound caused by tire traveling noise reaching the ear of the passenger;
A second filter representing a correlation between a noise cancellation sound output from the noise cancellation sound output speaker and a noise resulting from the noise cancellation sound reaching the passenger's ear,
The individual tire running noise signals collected by the tire running noise collecting microphone are passed through the first filters to generate individual tire running noise processing signals,
This individual tire running noise processing signal is added to generate an added signal after tire running noise processing,
The addition signal after the tire running noise processing is passed through the second filter to generate the noise canceling sound signal.
The vehicle noise reduction device according to claim 1, wherein the vehicle noise reduction device is provided. Virtual tire running noise generating means for generating virtual tire running noise as a virtual noise source;
Tire vibration means for vibrating each tire based on the virtual tire running noise generated by the virtual tire running noise generating means;
An ear microphone that picks up sound reaching the passenger's ear by virtue of the tire vibration means vibrating each tire; and
The signal processing means for generating the noise canceling sound signal is:
First delay means for generating a tire running noise delay signal by delaying a sound signal by the tire vibration means picked up by the tire running noise collecting microphone;
A first filter coefficient calculation unit that sets a coefficient of the first filter based on a difference between a sound signal obtained by the tire vibration means picked up by the ear microphone and the tire running noise delay signal;
The vehicle noise reduction device according to claim 3, wherein the vehicle noise reduction device is provided. Virtual noise canceling sound signal generating means as a virtual sound source;
Based on the virtual noise cancellation sound signal, the noise cancellation sound output speaker that outputs a virtual noise cancellation sound;
An ear microphone that picks up the virtual noise canceling sound reaching the passenger's ear, and
The signal processing means for generating the noise canceling sound signal is:
Second delay means for delaying the virtual noise cancellation sound signal to generate a virtual noise cancellation sound delay signal;
A second filter coefficient calculation unit that sets a coefficient of the second filter based on a difference between the virtual noise cancellation sound delayed signal and the virtual noise cancellation sound collected by the ear microphone;
The vehicle noise reduction device according to claim 3, wherein the vehicle noise reduction device is provided. The plurality of tire traveling noise collecting microphones are each attached to the vicinity of the holding portion of each tire that collects tire traveling noise.
The vehicle noise reduction device according to any one of claims 1 to 5, wherein: The noise canceling sound output speaker is disposed in the vicinity of the passenger's ear,
The vehicle noise reduction device according to any one of claims 1 to 6. The tire vibration means has a shape inserted between the tires and the ground.
The vehicle noise reduction device according to any one of claims 1 to 7, wherein the vehicle noise reduction device is provided. The signal processing means for generating the noise canceling sound signal is:
An individual audio signal filter in which a coefficient is set based on the audio signal sound of each channel output from the audio speaker picked up by the individual microphone for picking up tire running noise;
An individual audio signal subtracting unit that subtracts an output signal of the individual audio signal filter from the individual tire running noise signal, and
The output of the individual audio signal subtracting unit is used as an input of each first filter.
The vehicle noise reduction device according to claim 3 or 4, wherein the vehicle noise reduction device is provided. Virtual audio signal generation means for generating a virtual audio signal as a virtual sound source,
The audio signal filter is:
A coefficient is determined based on a difference between the virtual audio signal generated by the virtual audio signal generation unit and the virtual audio signal acquired by the individual tire traveling noise pickup microphone and the virtual audio signal. The
The vehicle noise reduction device according to claim 9. The signal processing means for generating the noise canceling sound signal is:
A plurality of specific frequency sound signal acquisition units for acquiring a sound signal of a specific frequency from each tire travel noise signal acquired by each of the tire travel noise sound pickup microphones;
A specific frequency sound signal subtracting unit that subtracts the specific frequency sound signal acquired by the specific frequency sound signal acquisition unit from each tire traveling noise signal acquired by each of the tire traveling noise sound collecting microphones,
The output of the specific frequency sound signal subtracting unit is used as the input of each first filter.
The vehicle noise reduction device according to claim 1, wherein the vehicle noise reduction device is provided. A plurality of microphones for collecting tire running noise, which individually collects the running noise generated by the tires of the vehicle and acquires a tire running noise signal;
Noise canceling sound for canceling the traveling noise of each tire reaching the ear of the passenger in the vehicle based on the noise canceling sound signal generated based on the traveling noise of the plurality of individually collected tires A noise canceling sound output speaker that outputs a noise cancellation method for a vehicle,
Correlation between individual tire vibration and sound caused by tire vibration reaching the passenger's ear, noise canceling sound output from the noise canceling sound output speaker, and noise canceling sound reaching the passenger's ear Generating the noise canceling sound signal based on the correlation with the sound caused by
A vehicle running noise reduction method characterized by the above. The vehicle further includes:
A plurality of first filters representing a correlation between tire traveling noise individually picked up by the plurality of tire traveling noise sound collecting microphones and sound caused by tire traveling noise reaching the ear of the passenger;
A second filter representing a correlation between a noise canceling sound output from the noise canceling sound output speaker and a sound resulting from the noise canceling sound reaching the passenger's ear;
Virtual tire running noise generating means for generating virtual tire running noise as a virtual noise source;
Tire vibration means for vibrating each tire based on the virtual tire running noise generated by the virtual tire running noise generating means;
An ear microphone that picks up sound reaching the passenger's ear by virtue of the tire vibration means vibrating each tire;
Virtual noise canceling sound signal generating means as a virtual sound source;
Based on the virtual noise cancellation sound signal, the noise cancellation sound output speaker that outputs a virtual noise cancellation sound;
An ear microphone that picks up the virtual noise canceling sound reaching the passenger's ear, and
Delaying a sound signal by the tire vibration means picked up by the tire traveling noise sound collecting microphone to generate a tire traveling noise delay signal;
Setting a coefficient of the first filter based on a difference between a sound signal from the tire excitation means picked up by the ear microphone and the tire running noise delay signal;
Delaying the virtual noise cancellation sound signal to generate a virtual noise cancellation sound delay signal;
Setting a coefficient of the second filter based on a difference between the virtual noise cancellation sound delayed signal and the virtual noise cancellation sound collected by the ear microphone;
The vehicle running noise reduction method according to claim 12, further comprising:

Images