State detection method, state detection system, and program

Abstract

The present disclosure relates to a state detection method, a state detection system, and a program capable of more reliably detecting a state of a vehicle. The state detection method according to the present disclosure calculates a tilt angle difference, which is a difference between tilt angles of each sensor, based on sensor values obtained from two or more sensors installed in a vehicle, and detects a state of the vehicle including the sensors based on the tilt angle difference. For example, the technology according to the present disclosure can be applied to a road noise cancellation system.

Description

Technical Field

[0001] This disclosure relates to condition detection methods, condition detection systems, and procedures, and more specifically, to condition detection methods, condition detection systems, and procedures capable of more reliably detecting the condition of a vehicle. Background Technology

[0002] When performing road noise cancellation using the sensor value of a triaxial accelerometer as a reference signal, the reference signal can change from its normal state due to sensor malfunctions (e.g., sensor failure or peeling, distortion of the vehicle body near the sensor, etc.). In this case, the output of the noise cancellation signal calculated using the reference signal also changes, and there is a possibility that the silencing effect at the listening position (e.g., ear position) may deteriorate due to the noise cancellation signal, or that the sound may increase at the ear position.

[0003] Therefore, there is a known technique that detects changes in the DC component of each reference signal when an abnormality or deterioration of each acceleration sensor installed near the wheels of a vehicle is detected (for example, see Patent Documents 1 and 2).

[0004] Reference List

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2019-82628

[0007] Patent Document 2: Japanese Patent Application Publication No. 2022-111594 Summary of the Invention

[0008] The problem to be solved by the present invention

[0009] In the technologies of Patent Documents 1 and 2, when the tilt angle of the vehicle body changes due to factors such as slope, there is a possibility that changes in the DC component of each reference signal could be detected as abnormal. On the other hand, when the threshold is set low to prevent such false detections, it is difficult to detect minor sensor malfunctions or stripping.

[0010] This disclosure is made in light of this situation, and its purpose is to enable more reliable detection of the vehicle's condition.

[0011] Solution to the problem

[0012] The state detection method disclosed herein includes: calculating a tilt angle difference based on sensor values ​​obtained from two or more sensors, the tilt angle difference being the difference between the tilt angles of the two or more sensors installed in the vehicle; and detecting the state of the vehicle, including the sensors, based on the tilt angle difference.

[0013] The state detection system disclosed herein is a state detection system comprising: a tilt difference calculation unit for calculating a tilt difference based on sensor values ​​obtained from two or more sensors, the tilt difference being the difference between the tilt angles of two or more sensors installed in the vehicle; and a state detection unit for detecting the state of the vehicle, including the sensors, based on the tilt difference.

[0014] The program disclosed herein is a program for causing a computer to perform the following processes: calculating a tilt angle difference based on sensor values ​​obtained from two or more sensors, the tilt angle difference being the difference between the tilt angles of the two or more sensors installed in the vehicle; and detecting the state of the vehicle, including the sensors, based on the tilt angle difference.

[0015] In this disclosure, a tilt angle difference is calculated based on sensor values ​​obtained from two or more sensors, the tilt angle difference being the difference between the tilt angles of the two or more sensors installed in the vehicle, and the state of the vehicle including the sensors is detected based on the tilt angle difference. Attached Figure Description

[0016] Figure 1 It is a diagram used to describe the calculation of the tilt angle.

[0017] Figure 2 This is a diagram illustrating a configuration example of a condition detection system applying the technology according to this disclosure.

[0018] Figure 3 It is a flowchart used to describe the process of state detection and processing.

[0019] Figure 4 This is a diagram illustrating a configuration example of a state detection system according to the first embodiment.

[0020] Figure 5 It is a flowchart used to describe the process of anomaly detection and handling.

[0021] Figure 6 This is a diagram illustrating an example of the state of an accelerometer during an anomaly.

[0022] Figure 7 It is a diagram showing the shift and difference in tilt angle of the accelerometer during an abnormal event.

[0023] Figure 8 This is a diagram illustrating an example of the state of an accelerometer under normal conditions.

[0024] Figure 9 This is a diagram showing the changes in the tilt angle and tilt angle difference of the accelerometer under normal conditions.

[0025] Figure 10 This is a view showing an example of an accelerometer sensor installation.

[0026] Figure 11 This is a diagram illustrating a configuration example of a state detection system according to the second embodiment.

[0027] Figure 12 It is a flowchart used to describe the process of distortion detection and processing.

[0028] Figure 13 This is a diagram illustrating a configuration example of a state detection system according to a third embodiment.

[0029] Figure 14 It is a flowchart used to describe the process of state detection and processing.

[0030] Figure 15 This is a block diagram illustrating an example of a computer's hardware configuration. Detailed Implementation

[0031] In the following text, a mode for carrying out this disclosure (hereinafter referred to as an implementation) will be described. Note that the description will be given in the following order.

[0032] 1. Based on the relevant technical issues and technical overview of this disclosure

[0033] 2. A condition monitoring system and its operation that utilizes the technology according to this disclosure.

[0034] 3. First Implementation Method (Detection of Sensor Anomalies)

[0035] 4. Second Implementation Method (Detection of Vehicle Body Deformation)

[0036] 5. Third Implementation Method (Sensor / Vehicle State Detection Using Estimation Model)

[0037] 6. Examples of computer hardware configurations

[0038] <1. Technical Issues and Technical Overview Based on This Disclosure>

[0039] (Related technical issues)

[0040] When road noise cancellation is performed using the sensor value of a triaxial accelerometer as a reference signal, the reference signal can change from its normal state due to sensor malfunctions (e.g., sensor failure or peeling, distortion of the vehicle body near the sensor, etc.). In this case, the output of the noise cancellation signal calculated using the reference signal also changes, and there is a possibility that the silencing effect at the listening position (e.g., ear position) may deteriorate due to the noise cancellation signal, or that the sound may increase at the ear position.

[0041] Therefore, there is a known technique that detects changes in the DC component of each reference signal when anomalies or deterioration of each acceleration sensor installed near the wheels of a vehicle are detected.

[0042] In this technology, when the vehicle's tilt angle changes due to factors such as slope, there is a possibility that changes in the DC component of each reference signal could be detected as an anomaly. On the other hand, when the threshold is set low to prevent such false detections, it is difficult to detect minor sensor malfunctions or stripping.

[0043] In addition, Japanese Patent Application Publication No. 2015-64347 (hereinafter referred to as Document 3) and Japanese Patent Application Publication No. 2021-4464 (hereinafter referred to as Document 4) disclose a technology in which an accelerometer is mounted in a static structure and detects attachment failure of the sensor or abnormality of the structure based on the tilt angle or tilt angle difference.

[0044] The application of the technology disclosed in Document 3 is limited to the structure. That is, when the difference in tilt angles of the accelerometers attached to the fixed surface increases, and a statistically significant difference occurs, i.e., the difference persists for a specified period, an anomaly determination is made; however, when a difference in tilt angles occurs, an immediate anomaly determination is not made. Therefore, it is difficult to apply the technology disclosed in Document 3 to the vehicle assumed in this disclosure. In addition, Document 3 does not consider usage scenarios such as vehicles where the tilt angle changes steadily due to driving, where installation errors of the accelerometers occur during factory manufacturing, or where small tilt angle differences continuously occur based on the tilt angle state when stationary. Therefore, it is difficult to set the probability density distribution used as a reference in Reference 3, and the technology disclosed in Document 3 cannot be applied to the above-mentioned usage scenarios. Furthermore, the sensor configuration of the technology disclosed in Document 4 is limited; like the technology disclosed in Document 3, it requires a certain amount of time for configuration before detection, and it is difficult to apply to situations where immediate detection of vehicle anomalies is desired.

[0045] Alternatively, one could consider using the correlation between sensors. However, since four independent vibrations are input to sensors installed near each wheel at, for example, four locations, the correlation between sensors is usually small, and it is difficult to use changes in correlation to determine anomalies.

[0046] (Based on the overview of the technology disclosed herein)

[0047] In the technology according to this disclosure, in the anomaly detection of a triaxial accelerometer used for onboard road noise cancellation (RNC), the difference in tilt angles calculated based on the DC component of the sensor value of each sensor is used, rather than each tilt angle calculated based on the DC component of each sensor.

[0048] In vehicle-specific environments where the tilt angle varies significantly even under normal conditions, if only the DC component and the tilt angle of each sensor are used as detection references, and if a strict threshold for anomaly detection is applied, there is a possibility of false detection due to factors such as slope.

[0049] On the other hand, in the technology according to this disclosure, by using the difference in tilt angle calculated for each sensor, no false detection occurs relative to the vehicle's tilt on a slope, and the threshold for anomaly detection can be strictly set. In other words, minor anomalies, sensor malfunctions, peeling, or deformation of the vehicle body near the sensor can be easily identified, unaffected by the vehicle's tilt, and anomalies in the RNC can be easily detected. As a result, by quickly detecting sensor anomalies in the sensor values ​​whose output becomes reference signals, RNC performance degradation can be prevented.

[0050] Furthermore, in the technology according to this disclosure, the acceleration sensor used in the RNC can be used as is, making it possible to detect abnormal attachment or peeling of the acceleration sensor, and also to detect distortion of the vehicle body without the need to add another sensor.

[0051] <2. A condition detection system and its operation that utilizes the technology according to this disclosure>

[0052] (introduce)

[0053] In a noise cancellation system applying the technology according to this disclosure, during normal driving, the tilt angle of each sensor is pre-calculated and maintained using the DC component of each axis based on sensor values ​​output from multiple triaxial acceleration sensors mounted in the vehicle body. Subsequently, if a change exists between the differences between the tilt angles calculated based on the tilt angles of the sensors during driving or when stationary, an anomaly is detected and the noise cancellation signal is shut off, thereby preventing accidental operation such as over-output in the RNC (Range Control Center).

[0054] Specifically, such as Figure 1 As shown, using sensor values ​​from three axes, the tilt angle under normal conditions or the difference in tilt angle between sensors (tilt angle difference) is pre-recorded from the acceleration sensor 2, which is installed near the bottom surface of the vehicle body (near the axle). In addition, multiple acceleration sensors 2 are installed on the vehicle 1.

[0055] The gravitational acceleration G is set such that G≈9.8 [m / s²] 2 Furthermore, the sensor values ​​(acceleration values) of the X, Y, and Z axes of accelerometer 2 are set to A. X A Y and A Z In the case of [the following], the following formula is used to calculate [the result]. Figure 1The tilt angle θ in the X-axis direction and the tilt angle ψ in the Y-axis direction are shown.

[0056] [Mathematical Expression 1]

[0057] [Mathematical Expression 2]

[0058] Here, the tilt angles θ and ψ are calculated from the DC component of the sensor value, and therefore it is necessary to obtain the DC component of the sensor value by applying a low-pass filter to the sensor value or by using the average value of the sensor value every few seconds.

[0059] During RNC execution, the tilt difference between sensors acquired from the accelerometer is calculated sequentially in a similar manner. An anomaly is detected if the calculated tilt difference exceeds a threshold for a certain period. The threshold can be a pre-set value, a value calculated from design drawings, a statistically set value, or a value set based on statistical analysis of time-series tilt difference data under normal conditions.

[0060] The statistical setpoint is a confidence interval for the probability distribution calculated based on the time series data of the dip angle difference under normal conditions, such as a 3δ interval. The threshold for the confidence interval based on the probability distribution can be set more precisely. For example, if there is an explanatory variable contributing to the dip angle difference (e.g., temperature), the dip angle difference may change with temperature. In this case, a regression model such as multiple regression analysis or Gaussian process regression can also be used to set the confidence interval in greater detail.

[0061] In addition, the threshold can be a value set based on the safety requirements specifications of the vehicle 1 design information or design drawings, which include the Young's modulus of the body material of the vehicle 1.

[0062] In a noise cancellation system employing the technology according to this disclosure, stopping the output of the noise cancellation signal upon detecting an anomaly is considered the simplest countermeasure. This allows for the immediate prevention of increased noise and unpleasant effects on the user caused by the abnormal noise cancellation signal. Furthermore, the result of the anomaly detection can be notified to another in-vehicle electronic device as vehicle information for the in-vehicle infotainment (IVI).

[0063] The technology disclosed herein is not limited to anomaly detection in the noise cancellation system described above, but can also be applied to the state detection of vehicle 1 or sensor 2, such as distortion detection of the vehicle body.

[0064] (Configuration of the condition monitoring system)

[0065] Figure 2This is a view showing a configuration example of a state detection system that applies the technology according to this disclosure.

[0066] Figure 2 The condition detection system 10 shown includes multiple sensors 2-1, 2-2, 2-3, ..., and 2-N, a tilt difference calculation unit 11, and a condition detection unit 12. The tilt difference calculation unit 11 and the condition detection unit 12 can be implemented in a single on-board electronic device installed on the vehicle 1, or they can be implemented in each of separate on-board electronic devices. Furthermore, the condition detection unit 12 may not be implemented in an on-board electronic device installed on the vehicle 1, but rather in a server such as the cloud.

[0067] Each of sensors 2-1 to 2-N is configured as an acceleration sensor mounted in vehicle 1. Sensors 2-1 to 2-N are configured as, for example, acceleration sensors for road noise cancellation (RNC) inside vehicle 1. In this case, sensors 2-1 to 2-N are attached near the wheels of vehicle 1. The attachment points of sensors 2-1 to 2-N are not limited to this, and sensors 2-1 to 2-N may be attached to any part of the vehicle body. Hereinafter, when sensors 2-1 to 2-N are not distinguished from each other, they are simply referred to as sensor 2.

[0068] The tilt difference calculation unit 11 calculates the tilt angle of each sensor 2 based on sensor values ​​(e.g., acceleration values) obtained from each sensor 2. Furthermore, the tilt difference calculation unit 11 calculates the tilt difference, which is the difference between the calculated tilt angles of the sensors 2. The tilt difference can be calculated as the difference between the tilt angles of two predetermined sensors 2. For example, the tilt difference can be calculated as the tilt difference between two adjacent sensors 2, or it can be calculated as the difference in tilt angle between a specific sensor 2 with high reliability and other sensors 2.

[0069] The state detection unit 12 detects the state of the vehicle 1, including the sensors 2, based on the tilt angle difference calculated by the tilt angle difference calculation unit 11. For example, if any of the tilt angle differences calculated by the tilt angle difference calculation unit 11 exceeds a threshold, the state detection unit 12 detects an anomaly in any of the sensors 2. In this case, the output of the noise cancellation signal in the noise cancellation system is stopped or limited. Furthermore, if any of the tilt angle differences calculated by the tilt angle difference calculation unit 11 exceeds a threshold, the state detection unit 12 detects distortion of the vehicle body near the sensors 2 used to calculate the tilt angle difference.

[0070] (Operation of the condition monitoring system)

[0071] Figure 3 This is a flowchart describing the state detection process performed by the state detection system 10.

[0072] In step S11, the tilt angle difference calculation unit 11 calculates the tilt angle of each sensor 2 based on the sensor value obtained from each of the multiple sensors 2 installed in the vehicle 1.

[0073] In step S12, the tilt difference calculation unit 11 calculates the tilt difference between the sensors 2 based on the calculated tilt angle of the sensor 2.

[0074] In step S13, the state detection unit 12 detects the state of the vehicle 1 containing each sensor 2 based on the tilt angle difference calculated by the tilt angle difference calculation unit 11.

[0075] For example, if any of the tilt differences calculated by the tilt difference calculation unit 11 exceeds a threshold, the state detection unit 12 detects an anomaly in any of the sensors 2 (specifically, the sensor 2 involved in the calculation of the tilt difference exceeding the threshold). In this case, the output of the noise cancellation signal in the noise cancellation system is stopped or limited. For example, if any of the tilt differences calculated by the tilt difference calculation unit 11 exceeds a threshold, the state detection unit 12 detects distortion of the vehicle body near the sensor 2 related to the tilt difference calculation. In this case, for example, the result of the distortion detection is notified to a predetermined vehicle body monitoring system.

[0076] Based on the above structure and processing, by using the difference in tilt angle calculated by each sensor, erroneous detections such as the vehicle's own tilt on the slope will not occur, and the threshold for detection can be strictly set, making it possible to detect the vehicle's state more reliably.

[0077] In the following, an implementation of a condition detection system based on the technology of this disclosure will be described.

[0078] <3. First Implementation Method (Sensor Anomaly Detection)>

[0079] Figure 4 This is a view illustrating a configuration example of a state detection system according to a first embodiment of the present disclosure.

[0080] exist Figure 4 In the state detection system 100 shown, with Figure 2 Components of the state detection system 10 are similar to those in the system and are represented by the same reference numerals, and their descriptions will be omitted as appropriate.

[0081] In other words, Figure 4 The status detection system 100 in the middle and Figure 2 The difference in the state detection system 10 is that a state detection unit 111 is set instead of a state detection unit 12, and a tilt difference storage unit 112 and a noise cancellation unit 113 are newly set.

[0082] In the condition detection system 100, each sensor 2 is configured as a triaxial acceleration sensor for the RNC inside the vehicle 1. Then, the tilt difference calculation unit 11 uses the DC components of the sensor values ​​(acceleration values) in the three axes (X, Y, Z axes) obtained from each sensor 2 to calculate the tilt angle of each sensor 2, thereby calculating the tilt difference between the sensors 2.

[0083] If any of the tilt angle differences calculated by the tilt angle difference calculation unit 11 exceeds the threshold, the state detection unit 111 detects an anomaly in any of the sensors 2 and outputs an anomaly detection signal indicating that an anomaly has been detected to the noise cancellation unit 113.

[0084] The tilt angle difference storage unit 112 stores the tilt angles of each sensor 2 calculated by the tilt angle difference calculation unit 11, and the tilt angle differences of each sensor 2 when the vehicle 1 is traveling under normal conditions. That is, the tilt angles and tilt angle differences under normal conditions stored in the tilt angle difference storage unit 112 are data collected in advance for the state detection unit 111 to statistically set the thresholds used to detect anomalies of each sensor 2. Alternatively, when using a preset value or a value calculated from a design drawing, etc., as the threshold, it is not necessary to collect the data under normal conditions in advance.

[0085] During vehicle 1 operation, noise cancellation unit 113 generates noise cancellation signals based on sensor values ​​from each sensor 2 and outputs the noise cancellation signals via speaker SP to perform RNC. Furthermore, in the event of an anomaly detected by state detection unit 111 in any of the sensors 2, that is, in response to the output of an anomaly detection signal from state detection unit 111, noise cancellation unit 113 stops or limits the output of the noise cancellation signals via speaker SP.

[0086] Reference Figure 5 The flowchart describes the process of anomaly detection and processing by the state detection system 100 on sensor 2. When vehicle 1 is moving, the noise cancellation process performed by noise cancellation unit 113 is executed... Figure 5 The processing.

[0087] In step S111, the tilt angle difference calculation unit 11 calculates the tilt angle of each sensor 2 based on the sensor value obtained from each of the multiple sensors 2 installed in the vehicle 1.

[0088] In step S112, the tilt difference calculation unit 11 calculates the tilt difference between the sensors 2 based on the calculated tilt angle of the sensors 2. The tilt difference calculation unit 11 calculates the tilt difference sequentially and provides the result to the state detection unit 111.

[0089] In step S113, the state detection unit 111 calculates the average value of the tilt angle difference within T seconds, which is calculated sequentially by the tilt angle difference calculation unit 11. In other words, the average value of the tilt angle difference is calculated every T seconds.

[0090] In step S114, the state detection unit 111 determines whether the average value of the tilt angle difference calculated every T seconds exceeds a threshold. If it is determined that the average value of the tilt angle difference does not exceed the threshold, the process returns to step S111, and the calculation of the average value of the tilt angle difference within a further T seconds is repeated.

[0091] On the other hand, if the average value of the tilt angle difference exceeds the threshold, the process proceeds to step S115, and the state detection unit 111 determines that any abnormality of the sensor 2 has been detected, generates an abnormality detection signal, and outputs the abnormality detection signal to the noise cancellation unit 113.

[0092] In step S116, the noise cancellation unit 113 stops or limits the output of the noise cancellation signal via the speaker SP. In other words, it performs a mute operation on the noise cancellation signal output from the speaker SP, or it performs an attenuation operation on the signal level of the noise cancellation signal.

[0093] Figure 6 This is a diagram illustrating an example of the state of an accelerometer during an anomaly.

[0094] Figure 6 A shows that sensors 2-1 and 2-2, attached to the moving vehicle 1, are in a normal state, and Figure 6 Figure B illustrates an abnormal state that occurs while the vehicle 1 is in motion and the sensor 2-2 is attached. Here, it is assumed that the tilt angle θ1 in the X-axis direction is calculated as the tilt angle of sensor 2-1, and the tilt angle θ2 in the X-axis direction is calculated as the tilt angle of sensor 2-2.

[0095] Figure 7 A represents in Figure 6 The example shows a view of the change in tilt angles θ1 and θ2 of sensors 2-1 and 2-2 over time, and Figure 7 B is a view representing the change in tilt angle difference θ1-θ2 over time. In this example, the average of tilt angles θ1 and θ2 is calculated per second, and the average of tilt angle difference θ1-θ2 is calculated per second. Furthermore, as... Figure 7 As shown in B, assuming that based on pre-collected data, the 3σ interval, in which 99.7% of the data falls under normal conditions, is set as the threshold for the tilt angle difference θ1-θ2. In other words, if the tilt angle difference θ1-θ2 continuously exceeds the 3σ interval, an anomaly is detected.

[0096] like Figure 7As shown in Figure A, until the past 8 seconds, sensors 2-1 and 2-2 were in normal condition, and there were no significant changes in tilt angles θ1 and θ2, respectively. However, after approximately 8 seconds, the installation status of sensor 2-2 became abnormal, and the tilt angle θ2 changed drastically.

[0097] like Figure 7 As shown in B, the tilt angle difference θ1-θ2 was within the 3σ interval until the past 8 seconds, but the tilt angle difference θ1-θ2 exceeded the 3σ interval because an attachment state anomaly occurred in sensor 2-2 in the past 8 seconds. Therefore, an anomaly was detected in sensor 2-2.

[0098] In addition, by storing the tilt angles of each sensor 2-1 and 2-2 as normal state data, it is easy to determine which sensor is malfunctioning when an anomaly is detected.

[0099] Figure 8 This is a diagram illustrating an example of the state of an accelerometer under normal conditions.

[0100] Figure 8 A shows that sensors 2-1 and 2-2 attached to the moving vehicle 1 are both in normal working order, and Figure 8 B shows the state where vehicle 1 is stopped on a slope, etc., and vehicle 1 itself is tilted in the state of A.

[0101] Figure 9 A represents the value of A in the context of the universe. Figure 8 The example shows a view of the change in tilt angles θ1 and θ2 of sensors 2-1 and 2-2 over time, and Figure 9 B is a view representing the change in tilt angle difference θ1-θ2 with respect to the passage of time.

[0102] Unlike static structures, vehicle 1 is a moving body, and therefore, the tilt angles θ1 and θ2 of sensors 2-1 and 2-2 themselves always fluctuate significantly.

[0103] In other words, such as Figure 9 As shown in Figure A, until the past 8 seconds, sensors 2-1 and 2-2 were in normal condition, and there were no significant changes in tilt angles θ1 and θ2, respectively. However, after approximately 8 seconds, vehicle 1 stopped on a slope, and the tilt angles θ1 and θ2 of sensors 2-1 and 2-2 changed dramatically.

[0104] Then, as Figure 9 As shown in B, the tilt angle difference θ1-θ2 falls within the 3σ interval until 8 seconds have passed. Furthermore, even after approximately 8 seconds have passed since vehicle 1 stopped on the slope, vehicle 1 itself tilts, and therefore, the tilt angle difference θ1-θ2 does not change significantly, and does not exceed the 3σ interval. That is, no anomaly was detected.

[0105] When setting thresholds based solely on the tilt angle of each sensor, false detections occur due to the large variations in each tilt angle caused by the vehicle's own tilt. However, when the vehicle's tilt changes, each sensor tilts in a similar manner, and therefore the change in the tilt angle difference between the sensors is smaller. This prevents false detections from the acceleration sensors under normal conditions.

[0106] Note that when the load on vehicle 1 changes, such as when vehicle 1 is in motion or when the load on vehicle 1 changes, the tilt angle difference between the sensors can be expected to change due to the temporary tilt of the vehicle body. However, the change in tilt angle difference in this case is relatively small compared to the change in tilt angle difference caused by sensor failure or stripping. Therefore, false detections can be avoided by appropriately setting the threshold based on the design drawings and data under normal conditions.

[0107] Based on the above structure and processing, by using the difference in tilt angle calculated for each sensor, erroneous detections such as the vehicle's own tilt on the slope are avoided, and the threshold for detection can be strictly set. Therefore, anomalies of each sensor can be detected more reliably.

[0108] <4. Second Implementation Method (Detection of Vehicle Body Deformation)>

[0109] Typically, it is ideal to attach the accelerometer for RNC near the axle, which serves as the vibration source. However, in a condition detection system applying the technology of this disclosure, the accelerometer is not limited to the bottom surface of the vehicle body 1, and can be attached to any part of the vehicle body 1. Any part of the vehicle body can include the ceiling, doors, and windows of the vehicle 1.

[0110] Figure 10 This is a view showing an example of an accelerometer sensor installation.

[0111] exist Figure 10 In the vehicle 1 shown, sensors 2-1 and 2-2 are attached to the bottom surface of the vehicle body, sensor 2-3 is attached to the windshield, sensors 2-4 and 2-5 are attached to the ceiling, and sensor 2-6 is attached to the rear window.

[0112] As described above, the accelerometer is installed at any part of the vehicle body 1, stores the tilt angle difference between sensors under normal conditions, and sequentially monitors changes in the tilt angle difference, enabling the detection of faults in the door section, distortion of the vehicle body, etc. Furthermore, by monitoring changes in the tilt angle difference, it is also possible to determine in which axial direction the distortion occurs.

[0113] In this embodiment, the sensor installed at any part of the vehicle body 1 is not limited to an acceleration sensor, as long as it can calculate the tilt angle difference. For example, a tilt sensor (angle sensor) that can detect tilt angle can be used.

[0114] Figure 11 This is a view illustrating a configuration example of a state detection system according to a second embodiment of the present disclosure.

[0115] exist Figure 11 In the state detection system 200 shown, with Figure 2 Components of the state detection system 10 are similar to those in the system and are represented by the same reference numerals, and their descriptions will be omitted as appropriate.

[0116] In other words, Figure 11 The status detection system 200 and Figure 2 The difference in the state detection system 10 is that a state detection unit 211 is set instead of a state detection unit 12, and a new tilt difference storage unit 212 is set.

[0117] In the condition detection system 200, each sensor 2 can be configured as a triaxial acceleration sensor for the RNC (Rotation Control Center) inside the vehicle 1, or it can be configured as a tilt sensor separate from the acceleration sensor for the RNC and capable of detecting tilt angle. Furthermore, as multiple sensors 2, the triaxial acceleration sensor and the tilt sensor can be mixed. Figure 11 In the state detection system 200, as referenced Figure 10 As described, in addition to the bottom surface of the vehicle body, the sensor 2 is attached to any part of the vehicle body, such as the ceiling, door or window of the vehicle body.

[0118] If any tilt angle difference calculated by the tilt angle difference calculation unit 11 exceeds the threshold, the state detection unit 211 detects the distortion of the vehicle body and outputs a distortion detection signal indicating that distortion has been detected.

[0119] The tilt angle difference storage unit 212 stores the tilt angles of each sensor 2 calculated by the tilt angle difference calculation unit 11, and the tilt angle differences between the sensors 2 in the normal state of the vehicle 1. In other words, the tilt angles and tilt angle differences in the normal state stored in the tilt angle difference storage unit 212 are pre-collected data so that the state detection unit 211 can statistically set a threshold for detecting distortion of the vehicle body. Alternatively, if a pre-set value or a value calculated from design drawings, etc., is used as the threshold, it is not necessary to pre-collect data in the normal state.

[0120] Reference Figure 12 The flowchart describes the process of distortion detection and processing of the vehicle body 1 by the state detection system 200. Figure 12The processing is not limited to being performed while vehicle 1 is in motion, and can be performed continuously.

[0121] In step S211, the tilt angle difference calculation unit 11 calculates the tilt angle of each sensor 2 based on the sensor value obtained from each of the plurality of sensors 2 attached to any part of the vehicle body.

[0122] In step S212, the tilt difference calculation unit 11 calculates the tilt difference between the sensors 2 based on the calculated tilt angle of the sensors 2. The tilt difference calculation unit 11 calculates the tilt difference sequentially and provides the result to the state detection unit 211.

[0123] In step S213, the state detection unit 211 calculates the average tilt angle difference over T seconds based on the tilt angle difference calculated sequentially by the tilt angle difference calculation unit 11. In other words, the average tilt angle difference is calculated every T seconds.

[0124] In step S214, the state detection unit 211 determines whether the average value of the tilt difference calculated every T seconds exceeds a threshold. If it is determined that the average value of the tilt difference does not exceed the threshold, the process returns to step S211, and the calculation of the average value of the tilt difference over a further T seconds is repeated.

[0125] On the other hand, if the average value of the tilt angle difference exceeds the threshold, the process proceeds to step S215, and the state detection unit 211 determines that distortion is detected in the body of the vehicle 1, generates a distortion detection signal, and outputs the distortion detection signal to, for example, a predetermined body monitoring system.

[0126] Based on the above configuration and processing, by using the difference in tilt angle calculated for each sensor, erroneous detections such as the vehicle tilting on a slope will not occur, and the threshold for detection can be strictly set, enabling more reliable detection of vehicle body distortion.

[0127] <5. Third Implementation Method (Sensor / Vehicle State Detection Using Estimation Models)>

[0128] Figure 13 This is a view showing a configuration example of a state detection system according to a third embodiment of the present disclosure.

[0129] exist Figure 13 In the state detection system 300 shown, with Figure 2 Components of the state detection system 10 are similar to those in the system and are represented by the same reference numerals, and their descriptions will be omitted as appropriate.

[0130] In other words, Figure 13 The status detection system 300 in the middle and Figure 2The difference in the state detection system 10 is that a learning unit 311 is newly set.

[0131] For example, learning unit 311 can be implemented on servers such as those in the cloud.

[0132] The learning unit 311 generates an estimation model for estimating the state of vehicle 1 or sensor 2 by learning the tilt angle difference corresponding to the state of vehicle 1 or sensor 2. For example, the learning unit 311 collects the tilt angle and tilt angle difference of each sensor 2 under normal and abnormal states through a wireless network or the like as data records for test vehicles and general vehicles, thereby generating an estimation model for estimating the pattern of each of the normal and abnormal states.

[0133] To generate the estimation model, in addition to general statistical methods such as t-tests and logistic regression analysis, deep learning can be applied. For example, methods that convert time-series data into spectrograms and use convolutional neural networks (CNNs) to estimate the spectrograms, or methods that use recurrent neural networks (RNNs) to estimate the state based on time-series data, can be employed. Furthermore, it is possible to learn not only from the normal changes in the tilt angle of vehicle 1 as input, but also from the tilt angle difference directly associated with abnormal states, thereby generating a high-precision estimation model.

[0134] The generated estimation model is used by the state detection unit 12 to detect the state of vehicle 1 or sensor 2.

[0135] Reference Figure 14 The flowchart describes the process of vehicle 1 condition detection processing, including sensor 2, performed by the condition detection system 300.

[0136] It is important to note that, in Figure 14 The processing in steps S311 and S312 of the flowchart is the same as that in... Figure 3 The processes in steps S11 and S12 of the flowchart are similar, therefore, their descriptions are omitted.

[0137] In other words, in step S313, the state detection unit 12 detects the state of the vehicle 1 including each sensor 2 by using the tilt angle and tilt angle difference calculated by the tilt angle difference calculation unit 11 as input and using the estimation model generated by the learning unit 311.

[0138] Based on the above processing, by using the difference in tilt angle calculated for each sensor to predict sensor failures and estimate the vehicle's condition, the difference in tilt angle can be used for repairs, safety alerts, maintenance, etc.

[0139] The foregoing has described an implementation in which the technology according to this disclosure is applied to a vehicle (automobile) equipped with two or more sensors. However, the technology according to this disclosure can also be applied to vehicles, such as railway vehicles and ships, as well as other moving bodies.

[0140] <6. Example of computer hardware configuration>

[0141] The above series of processes can be executed by hardware or by software. In the case where the processes are executed by software, the program configuring the software is installed from a program recording medium into a computer, general-purpose personal computer, or similar device integrated with dedicated hardware.

[0142] Figure 15 This is a block diagram illustrating an example configuration of computer hardware performing the aforementioned series of processes according to a program. At least some of the state detection systems 10, 100, 200, and 300 include, for example, having similar... Figure 15 The configuration of computer 500 shown is as described in the diagram.

[0143] The central processing unit (CPU) 501, read-only memory (ROM) 502 and random access memory (RAM) 503 are connected to each other via bus 504.

[0144] The input / output interface 505 is also connected to the bus 504. Input units 506, including a keyboard, mouse, etc., and output units 507, including a display, speakers, etc., are connected to the input / output interface 505. Furthermore, storage units 508, including hard disks, non-volatile memory, etc., communication units 509, including network interfaces, etc., and drivers 510 that drive removable media 511 are connected to the input / output interface 505.

[0145] In the computer 500 configured as described above, for example, the CPU 501 loads a program stored in the storage unit 508 into the RAM 503 via the input / output interface 505 and the bus 504 and executes the program, thereby performing the series of processes described above.

[0146] For example, a program executed by the CPU 501 may be provided by recording on a removable medium 511 or by using a wired or wireless transmission medium such as a local area network, the Internet, or digital broadcasting, and the program may be installed in a storage unit 508.

[0147] The program executed by the computer 500 may be a program that processes execution in a time sequence according to the order described in this specification, or it may be a program that processes execution in parallel or at necessary timed intervals, such as when a call is made.

[0148] Note that in this specification, "system" means a group of multiple components (devices, modules (parts), etc.), and it is irrelevant whether all components are housed in the same housing. Therefore, multiple devices and multiple modules housed in separate housings and connected via a network, and a single device housed in one housing, are all considered systems.

[0149] The effects described in this manual are merely examples and are not limited thereto; other effects can also be achieved.

[0150] The embodiments disclosed herein are not limited to those described above, and various modifications may be made without departing from the spirit of this disclosure.

[0151] For example, embodiments of this disclosure may have a cloud computing configuration, wherein multiple devices share a function and collaboratively perform processing via a network.

[0152] Furthermore, each step described in the above flowchart can be performed by a single device or can be performed by multiple devices in a shared manner.

[0153] Furthermore, in cases where a step includes multiple processes, the multiple processes included in a step can be performed by one device or by multiple devices in a shared manner.

[0154] The effects described in this manual are merely examples and are not limited thereto; other effects may be achieved.

[0155] Furthermore, the technology according to this disclosure may have the following configurations. (1)

[0157] A state detection method, comprising: The tilt angle difference is calculated based on sensor values ​​obtained from two or more sensors. The tilt angle difference is the difference between the tilt angles of the two or more sensors installed in the vehicle; and The vehicle's state is determined by sensors based on tilt difference detection. (2)

[0159] According to the state detection method in (1), where, The sensor is an acceleration sensor used for noise cancellation inside a vehicle, and The tilt angle of each of these sensors is calculated using the DC components of the sensor values ​​obtained from the sensors in the three axial directions. (3)

[0161] According to the state detection method in (2), where, If the tilt angle difference exceeds a threshold, detect any anomaly in the sensor, and If an anomaly is detected, stop or limit the output of the noise cancellation signal. (4)

[0163] According to the state detection method in (3), where, The signal level attenuation processing of the noise cancellation signal is used as a limitation for the noise cancellation signal. (5)

[0165] According to the state detection method in (3), where, The threshold is a preset value. (6)

[0167] According to the state detection method in (3), where, The threshold is a value set based on statistical analysis of time-series data of tilt angle difference under normal conditions. (7)

[0169] According to the state detection method in (3), where, The threshold is a value set based on safety requirements specifications. (8)

[0171] According to any one of the state detection methods in (1) to (6), where, The sensor is attached near the vehicle's wheels. (9)

[0173] According to the state detection method in (1), where, The sensor is attached to any part of the vehicle body, and Distortion of the vehicle body is detected based on tilt angle difference. (10)

[0175] According to the state detection method in (9), where, Any part of the vehicle body includes any one of the vehicle's ceiling, doors, and windows. (11)

[0177] According to the state detection method of (9) or (10), where, The sensors include accelerometers or tilt sensors. (12)

[0179] According to any one of the state detection methods in (1) to (11), where, An estimation model for estimating the state of a sensor or vehicle is generated by learning the tilt difference based on the state of the sensor or vehicle. (13)

[0181] According to the state detection method in (12), where, The state of a vehicle, including its sensors, is detected by using an estimation model, which uses the calculated tilt angle difference as input. (14)

[0183] A state detection system, comprising: The tilt difference calculation unit calculates the tilt difference based on sensor values ​​obtained from two or more sensors. The tilt difference is the difference between the tilt angles of the two or more sensors installed in the vehicle; and The state detection unit detects the state of the vehicle, including that of the sensors, based on tilt angle difference. (15)

[0185] A program for causing a computer to perform the following processes: The tilt angle difference is calculated based on sensor values ​​obtained from two or more sensors. The tilt angle difference is the difference between the tilt angles of the two or more sensors installed in the vehicle; and The vehicle's state is determined by sensors based on tilt difference detection.

[0186] Reference Symbol List

[0187] 1 vehicle

[0188] 2. Sensors 2-1 to 2-N

[0189] 11 Inclination Difference Calculation Unit

[0190] 12 Status Detection Units

[0191] 111 Status Detection Unit

[0192] 112 Tilt Difference Storage Units

[0193] 113 Noise Cancellation Unit

[0194] 211 Status Detection Unit

[0195] 212 Tilt Difference Storage Units

[0196] 311 Learning Unit.

Claims

1. A state detection method, comprising: The tilt angle difference is calculated based on sensor values ​​obtained from two or more sensors, wherein the tilt angle difference is the difference between the tilt angles of two or more sensors installed in the vehicle; as well as The tilt difference is used to detect the state of the vehicle, including the sensor.

2. The state detection method according to claim 1, wherein, The sensor is an acceleration sensor used for noise cancellation inside the vehicle. The tilt angle of each of the sensors is calculated using the DC components of the sensor values ​​obtained from the sensors in the three axial directions.

3. The state detection method according to claim 2, wherein, If the tilt angle difference exceeds a threshold, an anomaly is detected in any of the sensors, and If an anomaly is detected, stop or limit the output of the noise cancellation signal.

4. The state detection method according to claim 3, wherein, The signal level attenuation processing of the noise cancellation signal is used as a limitation of the noise cancellation signal.

5. The state detection method according to claim 3, wherein, The threshold is a preset value.

6. The state detection method according to claim 3, wherein, The threshold is a value set based on statistical analysis of the time series data of the tilt angle difference under normal conditions.

7. The state detection method according to claim 3, wherein, The threshold is a value set based on safety requirements specifications.

8. The state detection method according to claim 1, wherein, The sensor is attached near the vehicle's wheels.

9. The state detection method according to claim 1, wherein, The sensor is attached to any part of the vehicle body, and The distortion of the vehicle body is detected based on the tilt angle difference.

10. The state detection method according to claim 9, wherein, Any part of the vehicle body includes any one of the vehicle's ceiling, doors, and windows.

11. The state detection method according to claim 9, wherein, The sensors include accelerometers or tilt sensors.

12. The state detection method according to claim 1, wherein, An estimation model for estimating the state of the sensor or the vehicle is generated by learning the tilt difference based on the state of the sensor or the vehicle.

13. The state detection method according to claim 12, wherein, The state of the vehicle, including the sensors, is detected by using the estimation model, which uses the calculated tilt angle difference as input.

14. A state detection system, comprising: The tilt difference calculation unit calculates the tilt difference based on sensor values ​​obtained from two or more sensors, wherein the tilt difference is the difference between the tilt angles of two or more sensors installed in the vehicle; as well as A state detection unit detects the state of the vehicle, including the sensor, based on the tilt angle difference.

15. A program for causing a computer to perform the following processes: The tilt angle difference is calculated based on sensor values ​​obtained from two or more sensors, wherein the tilt angle difference is the difference between the tilt angles of the two or more sensors installed in the vehicle; and The tilt difference is used to detect the state of the vehicle, including the sensor.

Images