Vehicle anomaly detection system

Abstract

To provide an abnormality detection system for specifying an abnormality occurrence portion of a vehicle without using a temperature sensor.SOLUTION: A vehicle abnormality detection system is an abnormality detection system for a vehicle including a first sound collector, a second sound collector and a control device that detects an abnormality of the vehicle. The control device acquires first sound data from the first sound collector during traveling of the vehicle, acquires second sound data from the second sound collector, acquires first sound pressure data for each frequency from a first sound pressure waveform corresponding to the first sound data, acquires second sound pressure data for each of the frequencies from a second sound pressure waveform corresponding to the second sound data, detects an abnormality of the vehicle when determining both of first sound pressure change amount corresponding to the first sound pressure data and second sound pressure change amount corresponding to the second sound pressure data are threshold change amount or larger, and specifies a portion where the abnormality has occurred on the basis of a time difference between the first sound pressure waveform and the second sound pressure waveform.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a vehicle abnormality detection system.

Background Art

[0002] The exhaust device of an engine includes an exhaust manifold, an exhaust pipe, etc. The exhaust from the engine is collected by the exhaust manifold and discharged into the exhaust pipe. The exhaust contains harmful substances such as unburned gases (carbon monoxide and hydrocarbons), nitrogen oxides, and particulate matter. In order to remove these harmful substances in the exhaust, a catalyst device and a collection device are provided in the exhaust pipe as exhaust post-treatment devices.

[0003] The collection device for collecting particulate matter may be removed due to theft or vehicle modification. If the engine operates with the collection device removed, the emission amount of particulate matter discharged to the outside air may exceed the regulated value. For this reason, a technique for diagnosing the removal of the collection device has been proposed using a first exhaust temperature sensor that detects the temperature of the exhaust flowing into the collection device and a second exhaust temperature sensor that detects the temperature of the exhaust flowing out of the collection device (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, when detecting the removal of the collection device using a temperature sensor, there is a problem that it is difficult to accurately detect the removal of the collection device because the characteristic deviation due to the aging deterioration of the temperature sensor cannot be corrected.

[0006] Therefore, the present invention aims to provide an anomaly detection system that identifies the location of an anomaly in a vehicle without using a temperature sensor. [Means for solving the problem]

[0007] The vehicle abnormality detection system according to the present invention is a vehicle abnormality detection system comprising a first sound collector, a second sound collector, and a control device for detecting abnormalities in a vehicle, wherein the control device, while the vehicle is running, acquires first sound data from the first sound collector, acquires second sound data from the second sound collector, acquires first sound pressure data for each frequency from a first sound pressure waveform corresponding to the first sound data, acquires second sound pressure data for each frequency from a second sound pressure waveform corresponding to the second sound data, and if it determines that both the amount of change in first sound pressure corresponding to the first sound pressure data and the amount of change in second sound pressure corresponding to the second sound pressure data are greater than or equal to a threshold change, it detects an abnormality in the vehicle and identifies the location where the abnormality occurred based on the time difference between the first sound pressure waveform and the second sound pressure waveform.

[0008] In the above configuration, the control device may estimate the location of the abnormal noise generated in the vehicle based on the time difference, and if it estimates that the location of the noise is the tailpipe of the vehicle, it may identify the particulate matter collection device contained in the exhaust of the vehicle as the location of the noise.

[0009] In the above configuration, the control device may perform machine learning on the combination of the time difference, the first sound pressure data, and the second sound pressure data, and identify the location where the abnormality occurs based on the trained model obtained by machine learning on the combination.

[0010] In the above configuration, the control device may learn the initial sound pressure level of at least one of the first and second sound collectors, and if it determines that the sound pressure level has decreased relatively based on the learned model obtained by learning the sound pressure level, it may correct the sound pressure level of the one of the first and second sound collectors whose sound pressure level has decreased to return it to the initial sound pressure level.

[0011] In the above configuration, the first sound collector and the second sound collector may be positioned at different locations on the vehicle. [Effects of the Invention]

[0012] According to the present invention, it is possible to identify the location of a vehicle malfunction without using a temperature sensor. [Brief explanation of the drawing]

[0013] [Figure 1] This is an example of a vehicle's underside view. [Figure 2] This is a schematic diagram showing the configuration of the engine, ECU (Electronic Control Unit), and malfunction notification device. [Figure 3] This flowchart shows an example of the process performed by the ECU. [Figure 4] This is an example of an NV (Nuclear Value) determination map. [Figure 5] This diagram illustrates the detection of vehicle malfunctions. [Modes for carrying out the invention]

[0014] Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.

[0015] As shown in Figure 1, the engine 10 is located in the engine compartment at the front of the vehicle 100. The engine 10 may be a gasoline engine or a diesel engine. An exhaust system 20 is connected to the engine 10. The exhaust system 20 extends in the longitudinal direction of the vehicle 100.

[0016] The exhaust system 20 includes an exhaust manifold 21 and an exhaust pipe 22. The exhaust system 20 includes a catalytic converter 40 and a collection device 50 as exhaust aftertreatment devices. The exhaust system 20 includes a muffler 60 and a tailpipe 70. The catalytic converter 40, collection device 50, muffler 60 and tailpipe 70 are arranged in order from the front to the rear of the vehicle 100.

[0017] Note that when the engine 10 is a gasoline engine, the collection device 50 may be called a GPF (Gasoline Particulate Filter). When the engine 10 is a diesel engine, the collection device 50 may be called a DPF (Diesel Particulate Filter).

[0018] In the passenger compartment of the vehicle 100, a front first microphone 110, a front second microphone 120, a rear microphone 130, a steering wheel 140, etc. are provided. For example, the front first microphone 110 is arranged near the driver's seat 111. The front second microphone 120 is arranged near the passenger seat 121. The rear microphone 130 is arranged near the rear seat 131.

[0019] Note that the number of microphones is not limited to three, and may be two, or four or more. For example, any two of the front first microphone 110, the front second microphone 120, and the rear microphone 130 may be provided in the passenger compartment as the first microphone and the second microphone. Also, a different microphone from the front first microphone 110, the front second microphone 120, and the rear microphone 130 may be provided in the passenger compartment. Furthermore, the positions of the microphones may also be changed as appropriate. For example, the front first microphone 110, the front second microphone 120, and the rear microphone 130 may be arranged on the ceiling of the passenger compartment of the vehicle 100, or may be arranged on the door of the vehicle 100.

[0020] The front first microphone 110, the front second microphone 120, and the rear microphone 130 all include a microphone and collect various sounds generated in the vehicle 100. For example, the front first microphone 110, the front second microphone 120, and the rear microphone 130 collect the exhaust sound from the tail pipe 70. The front first microphone 110, the front second microphone 120, and the rear microphone 130 may also collect the sound generated by the collection device 50, the sound generated by the engine 10, etc.

[0021] Referring to FIG. 2, the details of the engine 10, the catalytic device 40, the collection device 50, the ECU 200, and the abnormality notification device 300 will be described. The ECU 200 is provided in the vehicle 100. For example, the ECU 200, the above-described front first microphone 110 (or front second microphone 120), and the rear microphone 130 can realize an abnormality detection system for the vehicle 100.

[0022] The engine 10 generates power for driving the vehicle 100 by burning the fuel injected from the fuel injection valve 12 inside each cylinder 11. In FIG. 2, for the purpose of preventing complication of the drawing, descriptions of the intake device, the ignition plug, etc. are omitted. Also, the fuel injection method is not limited to the direct injection type inside the cylinder, and a port injection type may be used. The exhaust device 20 purifies the exhaust generated inside each cylinder 11 and discharges it to the outside air.

[0023] The exhaust generated in each cylinder 11 of the engine 10 is collected by the exhaust manifold 2l and discharged to the exhaust pipe l2. The exhaust contains harmful substances such as unburned gas (carbon monoxide (CO) and hydrocarbon (HC)), nitrogen oxides (NOx), and particulate matter (PM (Particular Matter)). In order to remove such harmful substances in the exhaust, the catalytic device 40 and the collection device 50 are provided in the exhaust pipe 22 as an exhaust post-treatment device. A pressure sensor 55 is provided in the exhaust pipe 22 on the upstream side of the collection device 50. The pressure sensor 55 detects the exhaust pressure Pi near the inlet side of the collection device 50.

[0024] The catalytic device 40 includes a casing 41 and an exhaust purification catalyst 42 supported on a honeycomb-shaped carrier made of cordierite (ceramic) held inside the casing 41. The exhaust purification catalyst 42 is, for example, an oxidation catalyst (binary catalyst) or a three-way catalyst, and not limited to these, and an appropriate catalyst can be used according to the type and application of the engine 10. In the present embodiment, a three-way catalyst is used as the exhaust purification catalyst 42. When a three-way catalyst is used as the exhaust purification catalyst 42, the unburned gas and nitrogen oxides in the exhaust flowing into the catalytic device 40 are purified by the exhaust purification catalyst 42.

[0025] The collection device 50 is installed in the exhaust pipe 22 downstream of the catalyst device 40 in the exhaust flow direction. The collection device 50 comprises a casing 51 and a wall-flow type filter 52 held within the casing 51. The filter 52 collects particulate matter in the exhaust that flows into the collection device 50.

[0026] The ECU200 is an example of a control device that detects abnormalities in the vehicle 100, and comprises a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), input ports, and output ports. In other words, the ECU200 is a microcomputer. The CPU, RAM, ROM, input ports, and output ports are interconnected by a bidirectional bus. The ROM stores programs according to the flowchart described later, as well as, for example, map data.

[0027] The ECU200 receives various data from the pressure sensor 55, the front first sound collector 110, the front second sound collector 120, the rear sound collector 130, and the on-board sensors 150. The on-board sensors 150 include a GPS (Global Positioning System) sensor that determines the position data of the vehicle 100, a vehicle speed sensor that measures vehicle speed, and an outside temperature sensor. Based on the various data received, the ECU200 controls the abnormality notification device 300, which notifies the location of the abnormality.

[0028] For example, if the malfunction notification device 300 is a MIL (Malfunction Indication Lamp), the ECU 200 will illuminate the MIL to notify the driver of the location of the malfunction. If the malfunction notification device 300 is a navigation system, the ECU 200 will notify the driver of the location of the malfunction via voice guidance through the navigation system. The ECU 200 may also notify the driver of the location of the malfunction by sending an email to a mobile device they are carrying.

[0029] Next, the processes performed by the ECU200 will be explained with reference to Figures 3 to 5.

[0030] First, as shown in Figure 3, when the ECU 200 detects that the ignition switch (labeled IG in Figure 3) is turned ON (step S1), the ECU 200 determines whether or not the vehicle 100 is in motion (step S2). For example, the ECU 200 determines whether or not the vehicle speed of the vehicle 100 measured by the vehicle speed sensor is 0 km / h. If the vehicle 100 is not in motion (step S2: NO), the ECU 200 terminates the process.

[0031] If vehicle 100 is in motion (step S2: YES), ECU 200 acquires sound data (step S3). More specifically, ECU 200 acquires first sound data representing a sound pressure waveform from the front first sound collector 110 and second sound data representing a sound pressure waveform from the rear sound collector 130. Alternatively, ECU 200 may acquire first sound data from the front second sound collector 120. Alternatively, ECU 200 may acquire first sound data from the front first sound collector 110 and second sound data from the front second sound collector 120. Alternatively, ECU 200 may acquire first sound data from the front first sound collector 110, second sound data from the front second sound collector 120, and third sound data from the rear sound collector 130.

[0032] When sound data is acquired, the ECU200 acquires sound pressure data (step S4). More specifically, the ECU200 performs frequency analysis on the first sound data and the second sound data, respectively, and acquires first sound pressure data and second sound pressure data with frequencies proportional to the engine speed. Alternatively, the ECU200 may use a bandpass filter on each of the first and second sound data to acquire first sound pressure data and second sound pressure data with frequencies proportional to the engine speed. Alternatively, the ECU200 may perform frequency analysis on the first and second sound data, respectively, and acquire first sound pressure data and second sound pressure data with frequencies proportional to the combustion timing. Alternatively, the ECU200 may use a bandpass filter on each of the first and second sound data to acquire first sound pressure data and second sound pressure data with frequencies proportional to the combustion timing.

[0033] Upon acquiring sound pressure data, the ECU200 determines whether the absolute value of the sound pressure is greater than or equal to the first threshold (step S5). More specifically, the ECU200 determines whether both the absolute value of the first sound pressure in the first sound pressure data and the absolute value of the second sound pressure in the second sound pressure data are greater than or equal to the first threshold. The first threshold is a numerical value used to determine NV (Noise and Vibration), and as shown in Figure 4, it is set as an absolute value for each combination of engine speed (NE) and engine load factor (KL). In other words, the process in step S5 determines whether the noise and vibration are greater than or equal to the first threshold. Note that an NV judgment map in which the first threshold is set for each combination of engine speed and load factor is stored in the ECU200 for each vehicle 100.

[0034] If the absolute value of the sound pressure is less than the first threshold (Step S5: NO), the ECU 200 terminates processing. If the absolute value of the sound pressure is equal to or greater than the first threshold (Step S5: YES), the ECU 200 determines whether the amount of sound pressure change is equal to or greater than the second threshold (Step S6). More specifically, the ECU 200 determines whether both the amount of sound pressure change corresponding to the first sound pressure data and the amount of sound pressure change corresponding to the second sound pressure data are equal to or greater than the second threshold. For example, as shown in Figure 5, the ECU 200 is set to a threshold line L1 corresponding to the first threshold described above. The ECU 200 is also set to a threshold line L2 of the second threshold, which represents the absolute value of the amount of sound pressure change considering the deterioration of the front first sound collector 110, the front second sound collector 120, and the rear sound collector 130. Threshold line L2 is an example of a threshold change amount. When the collection device 50 is removed, the predetermined line L3, which represents the absolute value of the first sound pressure change based on the first sound pressure data, appears at or above the threshold line L2, regardless of the distance traveled. The same applies to the second sound pressure change based on the second sound pressure data. That is, when the collection device 50 is removed, the predetermined line L3 appears on a coordinate plane at or above the threshold lines L1 and L2, regardless of the distance traveled.

[0035] If the amount of sound pressure change is less than the second threshold (step S6: NO), the ECU 200 terminates processing. If the amount of sound pressure change is greater than or equal to the second threshold (step S6: YES), the ECU 200 detects an abnormality in the vehicle 100 (step S7). Abnormalities in the vehicle 100 include not only the removal of the collection device 50, but also damage to the brackets or stays (supports), and failure of any of the injectors provided in each cylinder 11 of the engine 10. In other words, the ECU 200 detects an abnormality in the vehicle 100 not only from abnormal noises caused by the removal of the collection device 50, but also from abnormal noises caused by damage to the brackets or stays.

[0036] If an abnormality is detected, the ECU 200 corrects the speed of sound (step S8). For example, the ECU 200 determines the altitude of the location where the vehicle 100 is traveling based on the position data obtained by the GPS sensor and the map data stored in the ECU 200. Then, based on the atmospheric pressure corresponding to the determined altitude and the ambient temperature measured by the ambient temperature sensor, the ECU 200 corrects a predetermined speed of sound, which is set in advance, to a faster or slower speed.

[0037] After correcting the speed of sound, the ECU 200 estimates the source of the abnormal noise based on the time difference of the sound pressure waveform corresponding to the corrected speed of sound (step S9). More specifically, the ECU 200 estimates the source of the abnormal noise based on the time difference between the first and second sound pressure waveforms corresponding to the corrected speed of sound. For example, if the collection device 50 is removed, the exhaust pulsation downstream of the collection device 50 increases. This causes the exhaust noise from the tailpipe 70 to increase. Therefore, if the ECU 200 identifies a loud exhaust noise from the tailpipe 70 based on the time difference between the first and second sound pressure waveforms corresponding to the corrected speed of sound, it estimates the source of the abnormal noise to be the tailpipe 70. On the other hand, if the ECU 200 identifies an abnormal noise from a source other than the tailpipe 70 based on the time difference between the first and second sound pressure waveforms corresponding to the corrected speed of sound, it estimates the source of the abnormal noise to be the bracket, injector, etc.

[0038] After estimating the location of the abnormal noise, the ECU 200 determines whether or not the noise originates from the tailpipe 70 (step S10). If the noise originates from a location other than the tailpipe 70 (step S10: NO), the ECU 200 terminates the process. If the noise originates from the tailpipe 70 (step S10: YES), the ECU 200 identifies the collection device 50 as the location of the abnormality (step S11). Once the collection device 50 is identified as the location of the abnormality, the ECU 200 notifies the abnormality notification device 300 of the abnormality (step S12) and terminates the process.

[0039] As described above, the abnormality detection system of vehicle 100 is equipped with multiple sound collectors, such as a front first sound collector 110, a front second sound collector 120, and a rear sound collector 130, and an ECU 200. The ECU 200 acquires, for example, first sound data from the front first sound collector 110 and second sound data from the rear sound collector 130 while vehicle 100 is in motion. When the ECU 200 acquires the first and second sound data, it acquires first sound pressure data for each frequency from the first sound pressure waveform corresponding to the first sound data, and second sound pressure data for each frequency from the second sound pressure waveform corresponding to the second sound data. When the ECU 200 acquires the first and second sound pressure data, it determines whether a predetermined line L3 representing the amount of change in the first sound pressure corresponding to the first sound pressure data and a predetermined line (not shown) representing the amount of change in the second sound pressure corresponding to the second sound pressure data are both above the threshold line L2. If the ECU200 determines that both the predetermined line L3 and the predetermined line (not shown) are above the threshold line L2, it detects an abnormality in the vehicle 100 and identifies the location of the abnormality based on the time difference between the first sound pressure waveform and the second sound pressure waveform.

[0040] This makes it possible to identify the location of an abnormality in the vehicle 100 without using a temperature sensor. For example, even without using a temperature sensor, it is conceivable that a sensor other than a temperature sensor could be added to the vehicle 100 to detect the removal of the collection device 50. However, securing a place to add a sensor is difficult, and if it were added under the exhaust pipe of the vehicle 100, other problems such as the sensor interfering with the road surface and being damaged could arise. According to this embodiment, the possibility of such other problems arising can also be eliminated.

[0041] Furthermore, the ECU 200 may improve the accuracy of anomaly detection in the vehicle 100 by utilizing the exhaust pressure detected by the pressure sensor 55. Specifically, there is a possibility that another vehicle (not shown) different from vehicle 100 may not be equipped with the sound collection device 50 due to theft or modification. If the ECU 200 performs anomaly detection based on the sound collection results from the tailpipe of such another vehicle, it may detect an anomaly in the other vehicle. Therefore, by utilizing both the exhaust sound from the tailpipe of vehicle 100 and the exhaust pressure detected by the pressure sensor 55, the ECU 200 can improve the accuracy of anomaly detection in vehicle 100.

[0042] In addition, the ECU 200 may use machine learning to acquire data on the time difference and the combination of the first and second sound pressure data, and then use this trained model to identify the location of the anomaly. This improves the accuracy of identifying the location of the anomaly. Furthermore, the ECU 200 uses machine learning to acquire data on the initial sound pressure level of at least one of the front first sound collector 110 and the rear sound collector 130. Based on the trained model acquired through machine learning of the sound pressure level, the ECU 200 may then determine that the sound pressure level has decreased relatively, and correct the sound pressure level of the one of the front first sound collector 110 and the rear sound collector 130 that has decreased to return it to its initial sound pressure level. This suppresses the decrease in anomaly detection accuracy caused by deterioration of the sound collectors.

[0043] Although preferred embodiments of the present invention have been described in detail above, the present invention is not limited to these specific embodiments, and various modifications and changes are possible within the scope of the gist of the invention as described in the claims. [Explanation of Symbols]

[0044] 10 Engines 50 Collection device 100 vehicles 110 Front No. 1 Sound Amplifier 120 Front 2nd Sound Amplifier 130 Rear sound amplifier 200 ECU

Claims

1. A vehicle abnormality detection system comprising a first sound collector, a second sound collector, and a control device for detecting abnormalities in the vehicle, The first sound collector and the second sound collector are positioned at different locations on the vehicle. The control device is While the vehicle is in motion, first sound data is acquired from the first sound collector, and second sound data is acquired from the second sound collector. First sound pressure data for each frequency is obtained from the first sound pressure waveform corresponding to the first sound data, and second sound pressure data for each frequency is obtained from the second sound pressure waveform corresponding to the second sound data. If it is determined that the first sound pressure change amount obtained by comparing the first sound pressure data acquired at different timings, and the second sound pressure change amount obtained by comparing the second sound pressure data acquired at different timings, are both greater than or equal to a threshold change amount, then an abnormality in the vehicle is detected. Based on the time difference between the first sound pressure waveform and the second sound pressure waveform, the location where the abnormality occurred is identified. A vehicle anomaly detection system characterized by the following:

2. The control device estimates the location of the abnormal noise generated in the vehicle based on the time difference, and if it estimates that the location of the noise is the vehicle's tailpipe, it identifies the particulate matter collection device contained in the vehicle's exhaust as the location of the noise. The vehicle abnormality detection system according to feature 1.

3. The control device performs machine learning on the combination of the time difference, the first sound pressure data, and the second sound pressure data, and identifies the location where the abnormality occurs based on the trained model obtained by machine learning on the combination. The vehicle abnormality detection system according to claim 1 or 2.

4. The control device learns the initial sound pressure level of at least one of the first and second sound collectors using machine learning, and if it determines that the sound pressure level has decreased relatively based on the learned model, it corrects the sound pressure level of the one of the first and second sound collectors whose sound pressure level has decreased to return it to the initial sound pressure level. The vehicle abnormality detection system according to claim 1 or 2.

Images