Determining the cause of a disturbing noise in a motor vehicle

Abstract

Method (200) with the following steps: - Detecting (205) a noise on board a motor vehicle (105) - determining (215) a driving state of the motor vehicle (105); and - determining (235) a possible cause of the noise based on a noise profile of the noise and the driving condition; characterized in - that a portion of the detected noise attributable to a sound system of the motor vehicle (105) is filtered out (220), - wherein sound signals provided by the sound system in the motor vehicle (105) are electrically sampled before their acoustic output.

Description

[0001] The invention relates to a technique for analyzing a motor vehicle. In particular, the invention relates to determining the cause of a disturbing noise on board a motor vehicle.

[0002] A motor vehicle can produce a variety of noises during operation, some of which are considered normal operating noises and others as noise. Noise can be perceived as unpleasant inside or outside the vehicle. Furthermore, the noise can indicate an existing or impending defect, which is why determining the cause of the noise is often difficult.

[0003] To determine the cause, a collection of typical noises and their associated causes can be compiled. A service technician can use this collection to better determine the cause of an occurring noise. Especially for a noise that occurs only sporadically, it can be difficult to find a suitable entry in the collection. The search is often complicated by the fact that only a description of the noise by a vehicle driver is available.

[0004] DE 10 2017 204 941 A1 relates to a method for noise detection on board a motor vehicle, whereby reference noises and minimum similarity values are defined.

[0005] DE 10 2017 200 372 A1 relates to a method for noise detection outside a motor vehicle.

[0006] DE 102 35 416 A1 relates to a method for detecting noise partly inside and partly outside the interior of a motor vehicle.

[0007] DE 10 2014 217 500 A1 concerns a system comprising a motor vehicle and a central component outside the vehicle.

[0008] DE 10 2007 051 261 A1 relates to a method according to the preamble of claim 1.

[0009] One object underlying the invention is to provide an improved technique for determining the cause of a noise on board a motor vehicle. The present invention achieves this object by means of the subject matter of the independent claims. Subclaims specify preferred embodiments.

[0010] According to a first aspect, a method comprises steps of detecting a noise on board a motor vehicle; determining a driving condition of the motor vehicle; and determining a possible cause of the noise based on a noise profile of the noise and the driving condition.

[0011] According to the invention, - that a portion of the recorded noise attributable to the vehicle's sound system is filtered out, - whereby sound signals provided by the sound system in the motor vehicle are electrically sampled before their acoustic output.

[0012] It has been recognized that noise is usually related to the driving condition, and the cause of the noise can be more accurately determined by considering the correlation between the vehicle condition and the noise. If a specific noise is sought, the driving condition can be varied until the noise occurs. This allows the noise to be reproduced more accurately, and a verbal or written description from an earwitness may no longer be necessary.

[0013] Further preferably, the noise profile is compared with a predetermined noise profile, which may in particular include frequency information. The noise profile preferably includes a spectrum, i.e., a discrete or continuous indication of the intensities of an acoustic signal at different frequencies or predetermined frequency ranges. In a further embodiment, the noise profile may also include a temporal progression of the signal intensity in a frequency range or a temporal progression of the spectrum.

[0014] The comparison can be made using a similarity measure that determines how well or to what extent one noise profile corresponds to another noise profile. To determine the similarity measure, a frequency distribution and / or a temporal progression of frequency intensities in the frequency profiles can be considered in various embodiments. The similarity measure can be standardized, for example, to a range between 0 and 1, where 0 indicates no similarity and 1 indicates identity between the two noise profiles.

[0015] If the similarity measure exceeds a predetermined threshold, a cause associated with the predetermined noise profile can be identified as a possible cause of the noise. In other words, the stored noise profile can be associated with a possible cause or fault condition, and if the noise profile of the detected noise matches the stored noise profile according to the similarity measure, an indication of the identified fault condition or possible cause can be output.

[0016] In one embodiment, the noise is captured using only a single microphone in the vehicle's interior. The microphone may already be present on board the vehicle for another reason, for example, as part of a hands-free system. Retrofitting the vehicle with hardware to implement the method may therefore be easy or even unnecessary.

[0017] In another variant, several microphones can be used so that, for example, on the basis of the stereo effect, a direction or position from which the sound is spreading can be determined by triangulation with regard to different recording times of the same sound at different microphone positions.

[0018] Preferably, a position on board the motor vehicle from which the noise is propagating is determined, with the cause being identified based on the position. For example, it can be determined that the noise is propagating from a left rear area of the motor vehicle and compared with predetermined noises that can occur in this area. Components that cannot exert any sound influence in this area can then no longer be considered as causes. To determine the position, a microphone arrangement can be used that enables localization, for example according to the stereo principle via phase or propagation time differences. Multiple microphones can be used and each microphone can be mounted either inside or outside the motor vehicle. The microphone preferentially registers airborne sound.It can also be considered that sound can propagate as structure-borne noise within the vehicle from a noise source and appear as airborne noise at a location distant from the actual source. Instead of a conventional airborne microphone, a structure-borne microphone can also be used for improved positioning.

[0019] In one embodiment, it is determined that the noise differs from a typical operating noise of the motor vehicle. Operating noises that are typical for the operation of the motor vehicle may be known. This avoids the search for a suspected defect in the motor vehicle when a noise known to be typical occurs.

[0020] In a further embodiment, it is determined that the cause differs from a normal operating procedure on board the motor vehicle. In other words, an indication of the determined possible cause can only be provided if it differs from a normal operating procedure. For example, a component whose operation or existence on board the motor vehicle is unknown to a driver may cause a noise that indicates a normal operating procedure. In this case, an indication of the known noise source or reason for the noise can be provided.

[0021] The driving state can include a rotational speed of a drive unit on board the motor vehicle. The drive unit can include, for example, a traction motor of the motor vehicle, but also, for example, a servo motor or an auxiliary drive for a unit of the motor vehicle. In this sense, a turbocharger can also be understood as a drive unit. Every rotating element can generate noises whose frequencies depend on the rotational speed of the element. Rotational speeds can be determined for numerous components or drive units on board the motor vehicle. For example, for a drive train with a drive motor and a transmission, the rotational speed of gears in the transmission can be determined from the rotational speed of the drive motor. Rotational speeds of wheels of the motor vehicle can be determined for each individual wheel using assigned rotational speed sensors.Based on its frequency, the noise can be attributed to a number of components on board the vehicle whose speeds are related to this frequency. This can also take into account the potential for harmonics to form. This can make it much easier to determine the cause based on the engine's speed.

[0022] The driving condition can include speed or acceleration. The speed or acceleration can act on the vehicle's chassis, and certain noises may only occur at certain speeds or accelerations. Longitudinal or lateral acceleration is preferably considered, and angular acceleration around the vertical axis can also be considered. The speed or acceleration can be measured or derived from other vehicle parameters. Such parameters can include, for example, brake pressure, torque demand, or steering angle.

[0023] The driving condition may include a configuration of the motor vehicle that influences its aerodynamics. Changes in aerodynamics can cause wind noise, which causes the noise. The configuration may include, for example, whether or to what extent a window, skylight, convertible top, or door of the motor vehicle is open. The presence of a special attachment, such as a roof rack, ski rack, bicycle rack, or trailer, may also be determined. The position of an adjustable element specifically designed to influence aerodynamics, such as an extendable spoiler or wing, may also be taken into account.

[0024] The detected noise can be preprocessed before the cause is determined. In one embodiment, a speech frequency band and / or, according to the invention, at least a portion attributable to a vehicle's sound system is filtered out of the detected noise or at least attenuated. For example, based on the noise, it can be determined whether a person on board the motor vehicle is speaking or not, whereby the acoustic conversation data can be neglected for determining the cause of the noise. A portion of the recorded sound attributable to the sound system, for example a hands-free device or a radio, can also be removed from further analysis. Filtering out can work particularly well if multiple microphones are installed on board the motor vehicle.For example, the microphone closest to a speech source may provide the clearest speech signal. This speech signal can then be filtered from the signals of one of the other microphones. In the case of a radio or other sound reinforcement system, the audio signal may already be present as a loudspeaker signal or as a source signal, allowing for even better source separation.

[0025] Preprocessing can also be performed depending on the driving condition. For example, if it is known that a roof rack is installed and the vehicle's speed is above a predetermined range, noise that typically occurs in this case can be filtered out of the detected noise. If the system is currently searching for noise or noise caused by the roof rack, such signals can also be amplified.

[0026] Since the motor vehicle may include a sound system, a portion of the noise attributable to the sound system is filtered out of the detected noise. The sound system may, for example, comprise a hands-free system or an audio or video entertainment system. According to the invention, sound signals provided by the sound system in the motor vehicle are electrically sampled before their acoustic output. This allows them to be preferably filtered out from microphone signals without having to detect them using another microphone.

[0027] In a further development of the method, a driving condition that provokes or influences the noise is determined based on the specific cause. A positive test can be performed by placing the motor vehicle in the specific driving condition and checking whether the expected noise occurs. The driving condition can include, in particular, a speed or maneuver, such as braking when negotiating a curve or a specific sequence of direction changes. The specific driving condition can be controlled automatically or by a driver.

[0028] According to a further aspect, a device on board a motor vehicle comprises a detection device configured to detect a noise; and a processing device configured to determine a driving state of the motor vehicle and to determine a possible cause of the noise based on a frequency of a noise profile of the noise and the driving state, and a sound device for providing sound signals in the motor vehicle.

[0029] According to the invention, the device is designed to - to filter out a portion of the recorded noise that can be attributed to the vehicle’s sound system and thereby - to electrically sample sound signals provided by the sound system in the motor vehicle before they are acoustically output.

[0030] The cause is determined on board the motor vehicle, so a wide variety of parameters or sensor values collected on board the vehicle can be used for the determination. Such values can be provided, in particular, via an interface that can lead to a data bus of the motor vehicle, for example, a CAN bus.

[0031] The device is preferably configured to carry out a method described herein, in whole or in part. For this purpose, the processing device may comprise a programmable microcomputer, and the method may be in the form of a computer program with program code means. The computer program product may also be stored on a computer-readable data carrier. Advantages or features of the method may be transferred to the device, and vice versa.

[0032] According to yet another aspect of the invention, a motor vehicle comprises a device as described herein.

[0033] According to yet another aspect of the invention, a system comprises a motor vehicle with a device described herein and a central component. The device comprises a first communication device, and the central component comprises a second communication device and a second processing device configured to determine a possible cause of the noise based on a frequency of the noise and the driving condition. The two communication devices are configured to transmit information to each other.

[0034] In this case, the cause is determined outside the vehicle, where greater processing power or a larger storage device for storing driving conditions or noise data may be more readily available. The central component can be configured to process information from multiple vehicles. The results of determinations can be transferred between different vehicles. The determination can be feedback-based, so that an improved second determination can be performed depending on whether the first determination is successful or unsuccessful.

[0035] In a further embodiment, the processing device is configured to anonymize the noise and / or the driving state before transmitting it to the central component. In particular, it can be ensured that no conclusions can be drawn about a specific motor vehicle, a specific driving situation, or a specific passenger in the motor vehicle based on the transmitted information.

[0036] The invention will now be described in more detail with reference to the accompanying drawings, in which: Fig. 1 a system; and Fig. 2 illustrates a flow chart of a process.

[0037] Fig. 1 shows a system 100 comprising a motor vehicle 105 with a device 110 and optionally a central component 115. The device 110 comprises a processing device 120 and at least one detection device 125 configured to detect a noise on board the motor vehicle 105. The detection device 125 may, in particular, comprise a microphone or a structure-borne sound sensor. Multiple detection devices 125, including of different designs, may be distributed at various locations on the motor vehicle 105. The detection device 125 may also be comprised by another system on board the motor vehicle 105, for example, a hands-free system or a diagnostic system.

[0038] Multiple detection devices 125 can also be grouped to allow the determination of a direction and / or position from which the noise is propagating. This determination can be performed, in particular, according to the stereo principle based on a phase shift between signals from the detection devices 125. For this purpose, the detection devices 125 can be arranged close to one another or at predetermined distances from one another.

[0039] Furthermore, the device 110 can be connected to an interface 130, which can allow communication with one or more systems or devices on board the motor vehicle 105. For example, an acceleration or speed value, a position of an actuator, a rotational speed of an assembly or a wheel of the motor vehicle 105, a configuration, a sound system, or the activity of another system on board the motor vehicle 105 can be obtained via the interface 130. Alternatively or in addition to the interface 130, one or more sensors can also be provided that can acquire information about a driving state of the motor vehicle, for example, configuration or movement information.

[0040] One or more simultaneously occurring values, each depending on a type of operation of the motor vehicle 105, can be summarized in an operating state of the motor vehicle 105.

[0041] The device 110 can also include a storage device 135, in which, in particular, information about a number of known noises can be stored, which, in particular, are not among the usual operating noises of the motor vehicle 105 and can occur, for example, when the motor vehicle 105 is subject to wear or a defect. Such noise information can be associated with one or more driving conditions in which the occurrence of the noise is possible or probable. Furthermore, one or more possible causes that enable or promote the occurrence of the respective noise in the respective driving conditions can be associated with the noise information or the driving condition.

[0042] The device 110 may further comprise a first communication device 140, which may be configured for wireless communication with a component outside the motor vehicle 105. A mobile radio network, for example, may be used for this purpose.

[0043] The central component 115 can comprise a second communication device 145, which is preferably configured to exchange information with the first communication device 140 of the device 110 on board the motor vehicle 105. This communication can be encrypted. Furthermore, the central component 115 can comprise a processing device 150 and optionally a storage device 155. The processing device 150 can operate in the same way as the processing device 120, and the storage device 155 can operate in the same way as the storage device 135, although generally more powerful components can also be used due to improved space or energy availability.

[0044] Fig. 2 shows a flowchart of a method 200. The method 200 can run in particular on the device 110 and / or the central component 115.

[0045] In a step 205, a noise on board the motor vehicle 105 is detected, preferably by means of the detection device 125. In a step 210, one or more sensors or the interface 130 can be scanned to obtain information that depends on a driving state of the motor vehicle 105. This information can, in particular, comprise driving dynamics, a configuration of the motor vehicle 105, or a state of an assembly or system on board the motor vehicle 105. In particular, the information can relate to a rotational speed of an element on board the motor vehicle 105. From at least one such piece of information, a driving state of the motor vehicle 105 can be formed in a step 215. For this purpose, detected information can be processed in a predetermined manner, for example by scaling, linking, integrating, or deriving.

[0046] In a step 220, the acquired noise information can be preprocessed. The goal of preprocessing is to extract characteristic components of the noise information as clearly as possible or to attenuate uncharacteristic components as much as possible. For example, preprocessing can include bandpass filtering, in which frequencies within the range of human speech are attenuated. This range can be approximately 300 Hz to 4 kHz. Other frequency ranges are also possible: for example, if children's voices are to be specifically filtered out, a higher frequency range can be selected. High-pass or low-pass filtering is also possible.

[0047] The preprocessing can be performed depending on the specific driving condition or at least one parameter influencing the driving condition. For example, if the speed of the motor vehicle 105 exceeds a predetermined threshold, a noise filter can be used to suppress wind noise. For example, if a window of the motor vehicle 105 is open, signals sampled near the window can be selectively attenuated. If noises originating on the exterior of the motor vehicle 105 are of particular interest, these signals can also be specifically amplified.

[0048] Based on the sampled noise information and the determined driving condition, a cause of the noise can then be determined. This determination can be made on board the motor vehicle 105 or by the central component 115. For this purpose, the provided information can be anonymized in a step 225 to make it more difficult to draw conclusions about the individual motor vehicle 105 or a person inside it from the information or a processing result thereof. In a step 230, the information can be transmitted from the device 110 to the central component 115, for which purpose an encrypted connection is preferably used.

[0049] After transmission in step 230 or if processing is performed by device 110, a cause for the detected noise can be sought in a step 235. To do so, it can first be checked whether the noise differs from a normal operating noise of motor vehicle 105. For this purpose, the driving state can already be taken into account. For example, during operation of motor vehicle 105, noise emissions from a drive motor are generally to be expected, wherein an intensity or frequency of the resulting noise should depend in a predetermined manner on a driving state of the motor vehicle, in particular a speed. The predetermined relationship can be provided, for example, in tabular form, as a characteristic curve, or as a function. If this relationship does not exist or does not exist in an expected manner, the detected noise is not a normal operating noise.

[0050] The noise can be analyzed according to its frequency, or more precisely according to a composition of the frequencies it contains. For this purpose, the noise information can be converted using a Fourier transform into a number of values or a characteristic curve that indicates such a composition. This characteristic curve can be called a noise profile. Intensities of one or more of the frequencies of the noise profile can then be compared with intensities of a predetermined noise profile stored in the storage device 135 or 155. In a further embodiment, a temporal progression of the frequency-dependent intensities of the noise is considered a noise profile and compared with corresponding information from a stored noise profile. In this case, the degree of similarity between the noise profiles can be determined in each case.

[0051] A stored noise can be assigned one or more possible causes, each of which can lead to the occurrence of the noise. These causes can also be stored in the storage device 135, 155. For example, a predetermined noise profile can be assigned to the cause of a worn wheel bearing. A cause can be assigned an observed frequency or probability of connection to the sampled noise. For example, a noise can be assigned a first cause with a high probability of occurrence and a second cause with a second probability of occurrence. This can express that the noise was observed in the past and / or in another motor vehicle more frequently due to the first cause than due to the second.The causes and probabilities can be collected on the basis of observations on a large number of motor vehicles, in particular of a comparable type.

[0052] For example, an operating noise classified as abnormal may be caused by a defective wheel bearing. When motor vehicle 105 travels around a curve, the outer wheel bearings are subjected to greater load than the inner ones. Furthermore, the outer wheels rotate faster than the inner wheels. For example, the noise from a left wheel bearing may become louder and assume higher frequencies when the motor vehicle travels around a right-hand bend, and become weaker and assume lower frequencies when it travels around a left-hand bend.

[0053] In particular, if the determination of a cause is inconclusive, for example, because the detected noise is consistent with wheel bearing damage on the left side of the motor vehicle 105, but the noise appears to be emanating from the right side of the vehicle, a driving condition or maneuver can be determined in a step 240 that should influence the detected noise in a predetermined manner if its cause has been correctly determined. An indication of such a driving condition can be output, preferably on board the motor vehicle 105. If processed by the central component 115, the indication can be transmitted appropriately. By maintaining predetermined time windows between the detection of a noise and the output of an indication of a specific cause or a suggested maneuver, the system 100 can be real-time capable. Reference symbol 100 systems 105 Motor vehicle 110 Device 115 central component 120 processing facility 125 scanning device 130 Interface 135 Storage device 140 first communication device 145 second communication device 150 processing facilities 155 Storage device 200 procedures 205 Detecting Sound 210 Scanning sensor or interface 215 Determine driving condition 220 Amplify / Filter 225 anonymize 230 transmit 235 Determine the cause 240 suggest test maneuvers

Claims

[1] Procedure (200) with the following steps: - Detecting (205) a noise on board a motor vehicle (105) - determining (215) a driving state of the motor vehicle (105); and - determining (235) a possible cause of the noise based on a noise profile of the noise and the driving condition; characterized by , - that a portion of the detected noise attributable to a sound system of the motor vehicle (105) is filtered out (220), - wherein sound signals provided by the sound system in the motor vehicle (105) are electrically sampled before their acoustic output. [2] The method (200) of claim 1, wherein the sampled sound signals are filtered out of microphone signals. [3] Method (200) according to claim 1 or 2, wherein the noise profile is compared with a predetermined noise profile, which in particular comprises frequency information; and wherein the comparison of the noise profile of the detected noise with the stored noise profile is carried out on the basis of a similarity measure which determines whether the profile of the detected noise corresponds to the stored noise profile. [4] The method (200) of claim 3, wherein the stored noise profile is associated with an error condition and, if the profile of the detected noise corresponds to the stored noise profile according to the similarity measure, an indication of the error condition is output. [5] Method (200) according to one of the preceding claims, wherein a position on board the motor vehicle (105) from which the noise propagates is determined (220), and wherein the cause is determined (235) on the basis of the position. [6] Method (200) according to one of the preceding claims, wherein it is determined - that the noise differs from a normal operating noise of the motor vehicle (105) and / or - that the cause differs from a normal operating procedure on board the motor vehicle (105). [7] Method (200) according to one of the preceding claims, wherein the driving state comprises a rotational speed of a drive unit on board the motor vehicle (105). [8] Method (200) according to one of the preceding claims, wherein a speech frequency band is filtered out (220) from the detected noise. [9] Method (200) according to one of the preceding claims, wherein a driving condition which provokes or influences the noise is determined (240) on the basis of the cause. [10] Device (110) on board a motor vehicle (105), comprising: - a detection device (125) configured to detect (205) a noise; and - a processing device (120) which is arranged to - to determine (235) a driving state of the motor vehicle (105) and - to determine a possible cause of the noise on the basis of a noise profile of the noise and the driving condition (235); and - a sound system for providing sound signals in the motor vehicle (105), wherein the device (110) is designed to - to filter out (220) from the detected noise a portion attributable to the sound system of the motor vehicle (105) and thereby - to electrically sample sound signals provided by the sound system in the motor vehicle (105) before they are acoustically output. [11] Motor vehicle (105) comprising a device (110) according to claim 10. [12] System (100), comprising - a motor vehicle (105) with a device (110) according to claim 10, wherein the device (110) comprises a first communication device (140); and - a central component (115) with - a second communication device (145) and - a second processing device (150) configured to determine a possible cause of the noise based on a frequency of the noise and the driving condition. [13] System (100) according to claim 12, wherein the processing device (120) is configured to anonymize (225) the noise and / or the driving state before transmission to the central component (115).

Images