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 that enables detecting of a state of a vehicle more reliably. A state detection method according to the present disclosure calculates, on the basis of sensor values obtained from two or more sensors installed in a vehicle, an inclination angle difference that is a difference between inclination angles of each of the sensors, and detects a state of the vehicle, including the sensors, on the basis of the inclination angle difference. The technology according to the present disclosure can be applied to a road noise cancelling system, for example.

Description

Condition detection method, condition detection system, and program

[0001] The present disclosure relates to a condition detection method, a condition detection system, and a program, and more particularly to a condition detection method, a condition detection system, and a program that enable more reliable detection of the condition of a vehicle.

[0002] When road noise canceling is performed using the sensor value of the three-axis acceleration sensor as a reference signal, the reference signal may change from its normal state due to an abnormality in the sensor itself, such as sensor failure or detachment, or 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, which may reduce the noise cancellation effect at a listening position such as the ear position, or may cause an increase in sound at the ear position due to the noise cancellation signal.

[0003] In response to this, there is known a technique for detecting a change in the DC component of each reference signal when detecting abnormalities or deterioration of acceleration sensors attached near the wheels of a vehicle (see, for example, Patent Documents 1 and 2).

[0004] JP 2019-82628 A JP 2022-111594 A

[0005] In the technologies disclosed in Patent Documents 1 and 2, when the inclination angle of the vehicle body changes on a slope, etc., there is a possibility that even a change in the DC component of each reference signal may be detected as an abnormality. On the other hand, if the threshold value is set too loosely to prevent such false detection, it becomes difficult to detect minute malfunctions or peeling of the sensor.

[0006] The present disclosure has been made in view of such circumstances, and aims to enable more reliable detection of the state of a vehicle.

[0007] The condition detection method disclosed herein is a condition detection method that calculates an inclination angle difference, which is the difference between the inclination angles of two or more sensors installed on a vehicle, based on sensor values ​​obtained from the sensors, and detects the condition of the vehicle including the sensors based on the inclination angle difference.

[0008] The condition detection system disclosed herein is a condition detection system that includes an inclination angle difference calculation unit that calculates an inclination angle difference, which is the difference between the inclination angles of two or more sensors installed on a vehicle, based on sensor values ​​obtained from the sensors, and a condition detection unit that detects the condition of the vehicle including the sensors based on the inclination angle difference.

[0009] The program disclosed herein is a program for causing a computer to execute a process of calculating an inclination angle difference, which is the difference between the inclination angles of two or more sensors installed on a vehicle, based on sensor values ​​obtained from the sensors, and detecting the state of the vehicle including the sensors based on the inclination angle difference.

[0010] In the present disclosure, an inclination angle difference, which is the difference between the inclination angles of two or more sensors installed in a vehicle, is calculated based on sensor values ​​obtained from the two or more sensors installed in the vehicle, and the state of the vehicle including the sensors is detected based on the inclination angle difference.

[0011] FIG. 1 is a diagram illustrating calculation of a tilt angle. FIG. 1 is a diagram illustrating an example configuration of a condition detection system to which the technology according to the present disclosure is applied. FIG. 2 is a flowchart illustrating the flow of a condition detection process. FIG. 2 is a diagram illustrating an example configuration of a condition detection system according to a first embodiment. FIG. 3 is a flowchart illustrating the flow of an abnormality detection process. FIG. 3 is a diagram illustrating an example state of an acceleration sensor when an abnormality occurs. FIG. 4 is a diagram illustrating the transition of the tilt angle and tilt angle difference of the acceleration sensor when an abnormality occurs. FIG. 4 is a diagram illustrating an example state of an acceleration sensor when normal. FIG. 5 is a diagram illustrating the transition of the tilt angle and tilt angle difference of the acceleration sensor when normal. FIG. 6 is a diagram illustrating an example installation of an acceleration sensor. FIG. 7 is a diagram illustrating an example configuration of a condition detection system according to a second embodiment. FIG. 8 is a flowchart illustrating the flow of a distortion detection process. FIG. 9 is a diagram illustrating an example configuration of a condition detection system according to a third embodiment. FIG. 10 is a flowchart illustrating the flow of a condition detection process. FIG. 11 is a block diagram illustrating an example configuration of computer hardware.

[0012] Modes for carrying out the present disclosure (hereinafter referred to as embodiments) will be described below in the following order.

[0013] 1. Problems with the prior art and an overview of the technology according to the present disclosure 2. A condition detection system to which the technology according to the present disclosure is applied and its operation 3. First embodiment (sensor abnormality detection) 4. Second embodiment (vehicle body distortion detection) 5. Third embodiment (sensor / vehicle condition detection using an estimation model) 6. Example of computer hardware configuration

[0014] 1. Problems with the Prior Art and Overview of the Technology Relating to the Present Disclosure> (Problems with the Prior Art) When performing road noise canceling using the sensor value of a three-axis acceleration sensor as a reference signal, the reference signal may change from its normal state due to an abnormality in the sensor itself, such as sensor failure or detachment, or 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, which could reduce the noise cancellation effect at a listening position such as ear position, or could cause increased sound at ear position due to the noise cancellation signal.

[0015] In response to this, a technique is known in which, when detecting abnormalities or deterioration in acceleration sensors attached near the wheels of a vehicle, changes in the DC components of each reference signal are detected.

[0016] With this technology, if the inclination angle of the vehicle body changes on a slope, etc., there is a possibility that even a change in the DC component of each reference signal may be detected as an abnormality. On the other hand, if the threshold is set too loosely to prevent such false detection, it becomes difficult to detect even a small malfunction or peeling of the sensor.

[0017] In addition, Japanese Patent Application Laid-Open No. 2015-64347 (hereinafter referred to as Document 3) and Japanese Patent Application Laid-Open No. 2021-4464 (hereinafter referred to as Document 4) disclose a technology in which an acceleration sensor is installed on a static structure and sensor installation failure or abnormality in the structure is detected based on the inclination angle or the difference in the inclination angle.

[0018] The technology disclosed in Literature 3 is limited to application to structures. Specifically, it determines an abnormality when a statistically significant difference appears as a large difference in the tilt angle of acceleration sensors attached to a fixed surface, i.e., when this difference continues for a predetermined period of time. However, it does not immediately determine an abnormality when a difference in tilt angle occurs, making it difficult to apply to vehicles, which are the technology of the present disclosure. Furthermore, Literature 3 does not anticipate use cases such as vehicles, where the tilt changes steadily as the vehicle moves, acceleration sensor installation errors occur during factory manufacturing, or slight tilt differences occur continuously depending on the tilt state when the vehicle is parked. Therefore, it is difficult to set a reference probability density distribution as anticipated in Literature 3, making it unapplicable. Furthermore, the technology disclosed in Literature 4 has limited sensor placement and, like the technology disclosed in Literature 3, requires a certain amount of time for detection, making it difficult to apply to cases where immediate detection of an abnormality in a vehicle is desired.

[0019] Another possible method is to use the correlation between sensors, but because four types of independent vibrations are input to sensors installed near each wheel, for example, in four locations, the correlation between the sensors is usually not large, making it difficult to determine abnormalities using changes in correlation.

[0020] (Summary of the technology disclosed herein) In the technology disclosed herein, when detecting an abnormality in a three-axis acceleration sensor used in in-vehicle road noise canceling (RNC), the difference in tilt angles calculated from the DC components of the sensor values ​​for each sensor is used, rather than the tilt angles calculated from the DC components of each sensor.

[0021] In a vehicle-specific environment where the inclination angle changes significantly even under normal circumstances, if the detection criteria are based solely on the DC component and inclination angle of each sensor, setting strict thresholds for abnormality detection could result in false detections on slopes, etc.

[0022] On the other hand, in the technology disclosed herein, by using the difference in inclination angle calculated for each sensor, false detection due to the inclination of the vehicle itself on a slope or the like does not occur, and a strict threshold for detecting an abnormality can be set. In other words, it is not affected by the inclination of the vehicle itself, and it becomes easier to determine minute abnormalities, failures, or peeling of the sensor, or distortion of the vehicle body near the sensor, making it easier to detect abnormalities in the RNC. Ultimately, by quickly detecting abnormalities in the sensor that outputs the sensor value that serves as the reference signal, it is possible to prevent degradation of the performance of the RNC.

[0023] Furthermore, in the technology disclosed herein, the acceleration sensor used in the RNC can be used as is, making it possible to detect installation abnormalities or detachment of the acceleration sensor, as well as distortion of the vehicle body, without adding any other new sensors.

[0024] 2. Status detection system employing the technology according to the present disclosure and its operation (Introduction) In a noise canceling system employing the technology according to the present disclosure, when the vehicle is traveling in a normal state, the tilt angle of each sensor is calculated using the DC component of each axis from the sensor values ​​output from multiple three-axis acceleration sensors installed on the vehicle body, and the calculated tilt angle is stored. Then, when the vehicle is traveling or stopped and a change occurs in the difference in tilt angle calculated from the tilt angles of the sensors, the system detects an abnormality and turns off the noise cancellation signal, thereby preventing unexpected operation such as excessive output in the RNC.

[0025] 1, the tilt angle in a normal state or the difference in tilt angle between the sensors (tilt angle difference) is recorded in advance using three-axis sensor values ​​from acceleration sensors 2 attached near the bottom of the body (near the axles) of the vehicle 1. It is assumed that a plurality of acceleration sensors 2 are attached to the vehicle 1.

[0026] Gravitational acceleration G≒9.8 [m / s ２ ], and the sensor values ​​(acceleration values) of the acceleration sensor 2 for each of the X, Y, and Z axes are A Ｘ , A Ｙ , A Ｚ In this case, the tilt angle θ in the X-axis direction and the tilt angle ψ in the Y-axis direction shown in FIG. 1 are calculated by the following equations.

[0027]

[0028]

[0029] Here, since the tilt angles θ and ψ are calculated from the DC component of the sensor value, 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.

[0030] During RNC execution, the inclination angle difference between the sensors acquired from the acceleration sensors is similarly calculated sequentially. If the calculated inclination angle difference continues to exceed a threshold value for a certain period of time, an abnormality is detected. The threshold value may be a preset value, a value calculated from design drawings, or a statistically set value, such as a value set based on statistical analysis of time-series data of inclination angle differences under normal conditions.

[0031] The statistically set value is the confidence interval of a probability distribution calculated from time-series data of the tilt angle difference under normal conditions, such as a 3δ interval. A threshold based on the confidence interval of the probability distribution can also be set more precisely. For example, if there is an explanatory variable (such as temperature) that contributes to the tilt angle difference, the tilt angle difference may change depending on the temperature. In this case, a more detailed confidence interval can be set using a regression model such as multiple regression analysis or Gaussian process regression.

[0032] The threshold value may also be a value set based on safety requirement specifications from design information and design drawings of the vehicle 1, including the Young's modulus of the body material of the vehicle 1.

[0033] In a noise canceling system incorporating the technology disclosed herein, when an abnormality is detected, the simplest response would be to stop outputting the noise canceling signal. This would immediately prevent increased noise caused by the abnormal noise canceling signal and any unpleasant effects on the user. Furthermore, the abnormality detection result may be notified to other in-vehicle electronic devices as vehicle information for the IVI (In-Vehicle Infotainment).

[0034] The technology according to the present disclosure is not limited to abnormality detection in the noise canceling system described above, but can also be applied to detecting the state of the vehicle 1 or the sensor 2, such as detecting distortion of the vehicle body.

[0035] (Configuration of Condition Detection System) FIG. 2 is a diagram showing an example configuration of a condition detection system to which the technology according to the present disclosure is applied.

[0036] 2 is configured to include a plurality of sensors 2-1, 2-2, 2-3, ..., 2-N, a tilt angle difference calculation unit 11, and a state detection unit 12. The tilt angle difference calculation unit 11 and the state detection unit 12 may be realized in a single on-board electronic device mounted on the vehicle 1, or may be realized in separate on-board electronic devices. Furthermore, the state detection unit 12 may be realized not in an on-board electronic device mounted on the vehicle 1, but in a server on a cloud or the like.

[0037] The sensors 2-1 to 2-N are each configured as an acceleration sensor installed in the vehicle 1. The sensors 2-1 to 2-N are configured as acceleration sensors used for road noise canceling (RNC) in the passenger compartment of the vehicle 1, for example. In this case, the sensors 2-1 to 2-N are attached near the wheels of the vehicle 1. However, the sensors 2-1 to 2-N may be attached at any location on the body of the vehicle 1. Hereinafter, when there is no need to distinguish between the sensors 2-1 to 2-N, they will simply be referred to as sensors 2.

[0038] The tilt angle 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 angle difference calculation unit 11 calculates a tilt angle difference, which is the difference between the calculated tilt angles of each sensor 2. The tilt angle difference may be calculated as the difference between the tilt angles of two predetermined sensors 2. For example, the tilt angle difference may be calculated as the difference between the tilt angles of two adjacent sensors 2, or may be calculated as the difference between the tilt angles of one specific sensor 2 with high reliability and another sensor 2.

[0039] The state detection unit 12 detects the state of the vehicle 1 including the sensors 2 based on the inclination angle differences calculated by the inclination angle difference calculation unit 11. For example, if any of the inclination angle differences calculated by the inclination angle difference calculation unit 11 exceeds a threshold, the state detection unit 12 detects an abnormality in any of the sensors 2. In this case, the output of a noise cancellation signal in the noise canceling system is stopped or limited. Furthermore, if any of the inclination angle differences calculated by the inclination angle difference calculation unit 11 exceeds a threshold, the state detection unit 12 detects distortion of the vehicle body near the sensor 2 used to calculate the inclination angle difference.

[0040] (Operation of the Status Detection System) FIG. 3 is a flowchart illustrating the flow of status detection processing by the status detection system 10. As shown in FIG.

[0041] In step S11 , the tilt angle difference calculation unit 11 calculates the tilt angle of each of the sensors 2 based on the sensor values ​​obtained from each of the multiple sensors 2 installed on the vehicle 1 .

[0042] In step S12 , the tilt angle difference calculation unit 11 calculates the tilt angle difference between the sensors 2 based on the calculated tilt angles of the sensors 2 .

[0043] In step S13 , the state detection unit 12 detects the state of the vehicle 1 including each sensor 2 based on the inclination angle difference calculated by the inclination angle difference calculation unit 11 .

[0044] For example, if any of the inclination angle differences calculated by the inclination angle difference calculation unit 11 exceeds a threshold, the state detection unit 12 detects an abnormality in any of the sensors 2 (specifically, the sensor 2 related to the calculation of the inclination angle difference that exceeds the threshold). In this case, the output of the noise cancellation signal in the noise canceling system is stopped or limited. Furthermore, for example, if any of the inclination angle differences calculated by the inclination angle 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 calculation of the inclination angle difference. In this case, for example, the result of the distortion detection is notified to a predetermined vehicle body monitoring system.

[0045] According to the above configuration and processing, by using the difference in inclination angle calculated for each sensor, false detection of the inclination of the vehicle itself on a slope, etc., does not occur, and a strict threshold value for detection can be set, making it possible to more reliably detect the state of the vehicle.

[0046] Hereinafter, an embodiment of a condition detection system to which the technology according to the present disclosure is applied will be described.

[0047] 3. First Embodiment (Sensor Abnormality Detection) FIG. 4 is a diagram illustrating an example of the configuration of a status detection system according to a first embodiment of the present disclosure.

[0048] In the condition detection system 100 shown in FIG. 4, the same components as those in the condition detection system 10 in FIG. 2 are denoted by the same reference numerals, and the description thereof will be omitted as appropriate.

[0049] That is, the status detection system 100 in FIG. 4 differs from the status detection system 10 in FIG. 2 in that it includes a status detection unit 111 instead of the status detection unit 12, and further includes a tilt angle difference storage unit 112 and a noise cancellation unit 113.

[0050] In the condition detection system 100, each sensor 2 is configured as a three-axis acceleration sensor used in an RNC in the cabin of the vehicle 1. The tilt angle difference calculation unit 11 calculates the tilt angle of each sensor 2 using the DC components of the sensor values ​​(acceleration values) in three-axis (X, Y, Z) directions obtained from each sensor 2, thereby calculating the tilt angle difference between the sensors 2.

[0051] If any of the inclination angle differences calculated by the inclination angle difference calculation unit 11 exceeds a threshold value, the state detection unit 111 detects an abnormality in any of the sensors 2 and outputs an abnormality detection signal indicating that an abnormality has been detected to the noise cancellation unit 113.

[0052] The inclination angle difference storage unit 112 stores the inclination angles of the sensors 2 and the inclination angle differences between the sensors 2 calculated by the inclination angle difference calculation unit 11 when the vehicle 1 is traveling in a normal state. That is, the inclination angles and inclination angle differences in a normal state stored in the inclination angle difference storage unit 112 are data collected in advance so that the state detection unit 111 can statistically set a threshold value for detecting an abnormality in any of the sensors 2. Note that if a preset value or a value calculated from design drawings or the like is used as the threshold value, data in a normal state does not need to be collected in advance.

[0053] The noise cancellation unit 113 executes RNC by generating a noise cancellation signal in accordance with the sensor values ​​from each sensor 2 and outputting the signal via the speaker SP while the vehicle 1 is traveling. Furthermore, when the status detection unit 111 detects an abnormality in any of the sensors 2, that is, when an abnormality detection signal is output from the status detection unit 111, the noise cancellation unit 113 stops or limits the output of the noise cancellation signal via the speaker SP.

[0054] The flow of the process of detecting an abnormality in the sensor 2 by the status detection system 100 will be described with reference to the flowchart in Fig. 5. The process in Fig. 5 is executed while the vehicle 1 is traveling and noise canceling is being performed by the noise canceling unit 113.

[0055] In step S111 , the tilt angle difference calculation unit 11 calculates the tilt angle of each of the sensors 2 based on the sensor values ​​obtained from each of the multiple sensors 2 installed on the vehicle 1 .

[0056] In step S112, the inclination angle difference calculation unit 11 calculates the inclination angle difference between the sensors 2 based on the calculated inclination angles of the sensors 2. The inclination angle difference calculation unit 11 sequentially calculates the inclination angle difference and supplies the result to the state detection unit 111.

[0057] In step S113, the state detection unit 111 calculates the average value of the tilt angle differences for T seconds, for the tilt angle differences sequentially calculated by the tilt angle difference calculation unit 11. That is, the average value of the tilt angle differences is calculated every T seconds.

[0058] In step S114, the state detection unit 111 determines whether the average value of the tilt angle differences calculated every T seconds exceeds the threshold value. If it is determined that the average value of the tilt angle differences does not exceed the threshold value, the process returns to step S111, and the calculation of the average value of the tilt angle differences for another T seconds is repeated.

[0059] On the other hand, if it is determined that the average value of the tilt angle differences exceeds the threshold value, the process proceeds to step S115, and the state detection unit 111 generates an abnormality detection signal, determining that an abnormality has been detected in one of the sensors 2, and outputs the signal to the noise cancellation unit 113.

[0060] In step S116, the noise cancellation unit 113 stops or limits the output of the noise cancellation signal via the speaker SP. That is, muting is performed on the noise cancellation signal output from the speaker SP, or attenuation is performed on the signal level of the noise cancellation signal.

[0061] FIG. 6 is a diagram showing an example of the state of the acceleration sensor when an abnormality occurs.

[0062] 6A shows a state in which sensors 2-1 and 2-2 attached to a moving vehicle 1 are both in a normal state, while FIG. 6B shows a state in which an abnormality occurs in the attachment state of sensor 2-2 while the vehicle 1 is moving. Here, it is assumed that the tilt angle of sensor 2-1 is calculated as tilt angle θ1 in the X-axis direction, and the tilt angle of sensor 2-2 is calculated as tilt angle θ2 in the X-axis direction.

[0063] 7A shows the time evolution of the inclination angles θ1 and θ2 of sensors 2-1 and 2-2, respectively, in the example of FIG. 6, and FIG. 7B shows the time evolution of the inclination angle difference θ1-θ2. In this example, the average values ​​of the inclination angles θ1 and θ2 are calculated every second, and the average value of the inclination angle difference θ1-θ2 is calculated every second. Furthermore, as shown in FIG. 7B, the threshold value for the inclination angle difference θ1-θ2 is set to a 3σ interval that contains 99.7% of the normal state data, based on data collected in advance. In other words, if the inclination angle difference θ1-θ2 continuously exceeds the 3σ interval, it is determined that an abnormality has been detected.

[0064] As shown in Figure 7A, up until 8 seconds have elapsed, both sensors 2-1 and 2-2 are in a normal state, and there is no significant change in their respective tilt angles θ1 and θ2. However, around 8 seconds after the time has elapsed, an abnormality occurs in the mounting condition of sensor 2-2, and its tilt angle θ2 changes significantly.

[0065] As shown in Figure B, the difference in tilt angle θ1-θ2 is within the 3σ interval until 8 seconds have elapsed, but around 8 seconds later, an abnormality occurs in the installation state of sensor 2-2, causing the difference in tilt angle θ1-θ2 to exceed the 3σ interval, thereby detecting an abnormality in sensor 2-2.

[0066] By storing the tilt angles of the sensors 2-1 and 2-2 as normal state data, it becomes possible to easily determine which sensor has experienced an abnormality when an abnormality is detected.

[0067] FIG. 8 is a diagram showing an example of the state of the acceleration sensor in a normal state.

[0068] Figure 8A shows a state in which sensors 2-1 and 2-2 attached to a moving vehicle 1 are both in a normal state, and Figure 8B shows a state in which the vehicle 1 has stopped on a slope or the like in the state shown in Figure 8A, and the vehicle 1 itself is tilted.

[0069] 9A is a diagram showing the transition of the inclination angles θ1 and θ2 of the sensors 2-1 and 2-2 over time in the example of FIG. 8, and FIG. 9B is a diagram showing the transition of the inclination angle difference θ1-θ2 over time.

[0070] Unlike a static structure, the vehicle 1 is a moving body, and therefore the inclination angles θ1 and θ2 of the sensors 2-1 and 2-2 themselves constantly fluctuate greatly.

[0071] That is, as shown in Figure 9A, until 8 seconds have elapsed, both sensors 2-1 and 2-2 are in a normal state, and there is no significant change in their respective inclination angles θ1 and θ2. However, around 8 seconds after the time has elapsed, the vehicle 1 stops on a slope or the like, and the inclination angles θ1 and θ2 of sensors 2-1 and 2-2 change significantly.

[0072] As shown in Figure B, the inclination angle difference θ1-θ2 remains within the 3σ interval until 8 seconds have elapsed. Furthermore, even after 8 seconds have elapsed since the vehicle 1 stopped on a slope or the like, the vehicle 1 itself remains inclined, so the inclination angle difference θ1-θ2 does not change significantly and does not exceed the 3σ interval. In other words, no abnormality is detected.

[0073] If thresholds are set by focusing on the inclination angle of each sensor, the change in each inclination angle is large, which can lead to false detection due to the inclination of the vehicle 1 itself, but when the inclination of the vehicle 1 itself changes, each sensor inclines in the same way, so the change in the difference in inclination angle between sensors is small. This makes it possible to prevent false detection by the acceleration sensor in a normal state.

[0074] It is possible that the difference in inclination angle between the sensors may change due to temporary deflection of the vehicle body when the load on the vehicle 1 changes, such as when the vehicle 1 is traveling or when the load on the vehicle 1 changes. However, the change in the difference in inclination angle in such a case is relatively small compared to the change in the difference in inclination angle due to sensor failure or peeling, and therefore, false detection can be avoided by appropriately setting the threshold value based on design drawings and data in a normal state.

[0075] According to the above configuration and processing, by using the difference in inclination angle calculated for each sensor, false detection of the inclination of the vehicle itself on a slope, etc., does not occur, and strict threshold values ​​for detection can be set, making it possible to more reliably detect abnormalities in each sensor.

[0076] 4. Second embodiment (detection of distortion of vehicle body) Typically, an acceleration sensor used in an RNC is ideally attached near the axle, which is a source of vibration. However, in a condition detection system to which the technology according to the present disclosure is applied, the acceleration sensor is not limited to being attached to the bottom surface of the vehicle body 1, but may be attached to any location on the vehicle body of the vehicle 1. The any location on the vehicle body may include any of the ceiling, doors, and windows of the vehicle 1.

[0077] FIG. 10 is a diagram showing an example of the installation of the acceleration sensor.

[0078] A vehicle 1 shown in FIG. 10 has sensors 2-1 and 2-2 attached to the bottom of the vehicle body, a sensor 2-3 attached to the front window, sensors 2-4 and 2-5 attached to the ceiling, and a sensor 2-6 attached to the rear window.

[0079] In this way, by installing acceleration sensors at any location on the body of the vehicle 1, storing the tilt angle difference between each sensor under normal conditions, and sequentially monitoring changes in the tilt angle difference, it becomes possible to detect malfunctions in the door section, distortion of the vehicle body, etc. Furthermore, by monitoring changes in the tilt angle difference, it becomes possible to determine in which axial direction the distortion is occurring.

[0080] In this embodiment, the sensor attached to any location on the body of vehicle 1 is not limited to an acceleration sensor, but may be any sensor capable of calculating the inclination angle difference, such as an inclination sensor (angle sensor) capable of detecting the inclination angle.

[0081] FIG. 11 is a diagram illustrating a configuration example of a status detection system according to the second embodiment of the present disclosure.

[0082] In the condition detection system 200 shown in FIG. 11, the same components as those in the condition detection system 10 in FIG. 2 are denoted by the same reference numerals, and the description thereof will be omitted as appropriate.

[0083] That is, the status detection system 200 in FIG. 11 differs from the status detection system 10 in FIG. 2 in that it includes a status detection unit 211 instead of the status detection unit 12 and further includes a tilt angle difference storage unit 212.

[0084] In the status detection system 200, each sensor 2 may be configured as a triaxial acceleration sensor used in an RNC in the cabin of the vehicle 1, or as an inclination sensor capable of detecting an inclination angle that is provided separately from the acceleration sensor used in the RNC. Furthermore, a combination of triaxial acceleration sensors and inclination sensors may be used as the multiple sensors 2. In the status detection system 200 of Fig. 11 , the sensors 2 may be attached to any location on the vehicle body, such as the bottom surface of the vehicle body of the vehicle 1, as well as the ceiling, doors, and windows of the vehicle 1, as described with reference to Fig. 10 .

[0085] If any of the inclination angle differences calculated by the inclination angle difference calculation unit 11 exceeds a threshold value, the state detection unit 211 detects distortion of the vehicle body and outputs a distortion detection signal indicating that distortion has been detected.

[0086] The inclination angle difference storage unit 212 stores the inclination angles of the sensors 2 and the inclination angle differences between the sensors 2 calculated by the inclination angle difference calculation unit 11 when the vehicle 1 is in a normal state. That is, the inclination angles and inclination angle differences in the normal state stored in the inclination angle difference storage unit 212 are data collected in advance so that the state detection unit 211 can statistically set a threshold value for detecting distortion of the vehicle body. Note that, if a preset value or a value calculated from design drawings or the like is used as the threshold value, data in the normal state does not need to be collected in advance.

[0087] The flow of the process of detecting distortion of the body of the vehicle 1 by the status detection system 200 will be described with reference to the flowchart of Fig. 12. The process of Fig. 12 may be executed at all times, not only while the vehicle 1 is traveling.

[0088] In step S211, the tilt angle difference calculation unit 11 calculates the tilt angle of each of the sensors 2 based on the sensor values ​​obtained from each of the multiple sensors 2 attached to any location on the body of the vehicle.

[0089] In step S212, the inclination angle difference calculation unit 11 calculates the inclination angle difference between the sensors 2 based on the calculated inclination angles of the sensors 2. The inclination angle difference calculation unit 11 sequentially calculates the inclination angle difference and supplies the result to the state detection unit 211.

[0090] In step S213, the state detection unit 211 calculates the average value of the tilt angle differences for T seconds from the tilt angle differences sequentially calculated by the tilt angle difference calculation unit 11. That is, the average value of the tilt angle differences is calculated every T seconds.

[0091] In step S214, the state detection unit 211 determines whether the average value of the tilt angle differences calculated every T seconds exceeds the threshold value. If it is determined that the average value of the tilt angle differences does not exceed the threshold value, the process returns to step S211, and the calculation of the average value of the tilt angle differences for another T seconds is repeated.

[0092] On the other hand, if it is determined that the average value of the inclination angle difference exceeds the threshold value, the process proceeds to step S215, and the state detection unit 211 generates a distortion detection signal, determining that distortion has been detected in the body of the vehicle 1, and outputs the signal to, for example, a predetermined body monitoring system.

[0093] According to the above configuration and processing, by using the difference in inclination angle calculated for each sensor, false detection of the inclination of the vehicle itself on a slope, etc., does not occur, and a strict threshold value for detection can be set, making it possible to more reliably detect distortion of the vehicle body.

[0094] 5. Third Embodiment (Sensor / Vehicle State Detection Using Estimation Model) FIG. 13 is a diagram illustrating an example of the configuration of a state detection system according to a third embodiment of the present disclosure.

[0095] In the condition detection system 300 shown in FIG. 13, the same components as those in the condition detection system 10 in FIG. 2 are denoted by the same reference numerals, and the description thereof will be omitted as appropriate.

[0096] That is, the condition detection system 300 in FIG. 13 differs from the condition detection system 10 in FIG. 2 in that a learning unit 311 is newly provided.

[0097] The learning unit 311 is realized, for example, on a server on the cloud.

[0098] The learning unit 311 generates an estimation model that estimates the state of the vehicle 1 or the sensor 2 by learning the inclination angle difference according to the state of the vehicle 1 or the sensor 2. For example, the learning unit 311 generates an estimation model that estimates the patterns of the normal state and the abnormal state by collecting the inclination angle and the inclination angle difference of each sensor 2 in normal and abnormal states as a data log of a test vehicle or a general vehicle via a wireless network or the like.

[0099] To generate the estimation model, general statistical methods such as t-tests and logistic regression analysis can be used, as well as deep learning, such as a method of converting time-series data into a spectrogram and estimating it with a CNN (Convolutional Neural Network), or a method of estimating the state based on time-series data using an RNN (Recurrent Neural Network).A highly accurate estimation model can be generated by learning not only changes in the inclination angle that normally occur in the vehicle 1, but also inclination angle differences that are directly related to abnormal states as inputs.

[0100] The generated estimation model is used by the state detection unit 12 to detect the state of the vehicle 1 or the sensor 2 .

[0101] The flow of the state detection process of the vehicle 1 including the sensor 2 by the state detection system 300 will be described with reference to the flowchart of FIG.

[0102] The processes in steps S311 and S312 in the flowchart of FIG. 14 are the same as the processes in steps S11 and S12 in the flowchart of FIG. 3, and therefore a description thereof will be omitted.

[0103] That is, in step S313, the state detection unit 12 receives the inclination angle and inclination angle difference of each sensor 2 calculated by the inclination angle difference calculation unit 11 as input, and detects the state of the vehicle 1 including each sensor 2 by using the estimation model generated by the learning unit 311.

[0104] According to the above processing, the difference in tilt angle calculated for each sensor can be used to predict sensor failure and estimate the vehicle state, which can be used for repairs, safety warnings, maintenance, etc.

[0105] The above describes an embodiment in which the technology disclosed herein is applied to a vehicle (automobile) equipped with two or more sensors, but it can also be applied to vehicles such as railway cars and ships, and other moving bodies.

[0106] 6. Example of Computer Hardware Configuration The above-described series of processes can be executed by hardware or software. When the series of processes is executed by software, the program constituting the software is installed from a program recording medium into a computer incorporated in dedicated hardware, a general-purpose personal computer, or the like.

[0107] 15 is a block diagram showing an example of the hardware configuration of a computer that executes the above-described series of processes using a program. At least a part of the condition detection systems 10, 100, 200, and 300 is configured by, for example, a computer 500 having a configuration similar to that shown in FIG.

[0108] A CPU (Central Processing Unit) 501 , a ROM (Read Only Memory) 502 , and a RAM (Random Access Memory) 503 are interconnected by a bus 504 .

[0109] An input / output interface 505 is also connected to the bus 504. An input unit 506 including a keyboard, a mouse, etc., and an output unit 507 including a display, a speaker, etc. are connected to the input / output interface 505. Also connected to the input / output interface 505 are a storage unit 508 including a hard disk, a nonvolatile memory, etc., a communication unit 509 including a network interface, etc., and a drive 510 that drives removable media 511.

[0110] In the computer 500 configured as described above, the CPU 501 performs the above-described series of processes by, for example, loading a program stored in the memory unit 508 into the RAM 503 via the input / output interface 505 and the bus 504 and executing it.

[0111] The program executed by the CPU 501 is installed in the storage unit 508 by being recorded on, for example, a removable medium 511 or provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital broadcasting.

[0112] The program executed by computer 500 may be a program that processes in chronological order according to the order described in this specification, or may be a program that processes in parallel or at the required timing, such as when called.

[0113] In this specification, a system refers to a collection of multiple components (devices, modules (components), etc.), regardless of whether all of the components are contained in the same housing. Therefore, multiple devices housed in separate housings and connected via a network, and a single device housed in a single housing with multiple modules, are both systems.

[0114] The effects described in this specification are merely examples and are not limiting, and other effects may also be present.

[0115] The embodiments of the present disclosure are not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure.

[0116] For example, the embodiment of the present disclosure can be configured as a cloud computing system in which a single function is shared and processed collaboratively by multiple devices via a network.

[0117] Furthermore, each step described in the above flowchart can be executed by one device, or can be shared and executed by a plurality of devices.

[0118] Furthermore, when one step includes multiple processes, the multiple processes included in that one step can be executed by one device or can be shared and executed by multiple devices.

[0119] The effects described in this specification are merely examples and are not limiting, and other effects may also be present.

[0120] Furthermore, the technology according to the present disclosure may be configured as follows. (1) A condition detection method for calculating, based on sensor values ​​obtained from two or more sensors installed in a vehicle, a tilt angle difference that is a difference between the tilt angles of the sensors, and detecting a condition of the vehicle including the sensors based on the tilt angle difference. (2) The condition detection method according to (1), wherein the sensors are acceleration sensors used for noise canceling in the passenger compartment of the vehicle, and the tilt angle of each of the sensors is calculated using DC components of the sensor values ​​in three axial directions obtained from the sensors. (3) The condition detection method according to (2), wherein, if the tilt angle difference exceeds a threshold, an abnormality in any of the sensors is detected, and if an abnormality is detected, output of a noise cancellation signal is stopped or limited. (4) The condition detection method according to (3), wherein, as a limit to the noise cancellation signal, a signal level of the noise cancellation signal is attenuated. (5) The condition detection method according to (3), wherein the threshold is a preset value. (6) The condition detection method according to (3), in which the threshold value is a value set based on statistical analysis of time-series data of the inclination angle difference under normal conditions. (7) The condition detection method according to (3), in which the threshold value is a value set based on safety requirement specifications. (8) The condition detection method according to any of (1) to (6), in which the sensor is attached near a wheel of the vehicle. (9) The condition detection method according to (1), in which the sensor is attached to an arbitrary location on a body of the vehicle and detects distortion of the body based on the inclination angle difference. (10) The condition detection method according to (9), in which the arbitrary location on the body includes any of a ceiling, a door, and a window of the vehicle. (11) The condition detection method according to (9) or (10), in which the sensor includes an acceleration sensor or an inclination sensor. (12) The condition detection method according to any of (1) to (11), in which an estimation model for estimating the state of the sensor or the vehicle is generated by learning the inclination angle difference according to the state of the sensor or the vehicle.(13) A condition detection method according to (12), which detects a condition of the vehicle including the sensor by using the estimation model with the calculated inclination angle difference as an input. (14) A condition detection system including: an inclination angle difference calculation unit that calculates an inclination angle difference that is a difference between the inclination angles of two or more sensors installed on a vehicle based on sensor values ​​obtained from the sensors, and a condition detection unit that detects the condition of the vehicle including the sensor based on the inclination angle difference. (15) A program that causes a computer to execute a process of calculating an inclination angle difference that is a difference between the inclination angles of the two or more sensors installed on a vehicle based on sensor values ​​obtained from the two or more sensors installed on the vehicle, and detecting the condition of the vehicle including the sensor based on the inclination angle difference.

[0121] REFERENCE SIGNS LIST 1 vehicle, 2, 2-1 to 2-N sensors, 11 inclination angle difference calculation unit, 12 state detection unit, 111 state detection unit, 112 inclination angle difference storage unit, 113 noise cancellation unit, 211 state detection unit, 212 inclination angle difference storage unit, 311 learning unit

Claims

1. A condition detection method that calculates an inclination angle difference, which is the difference between the inclination angles of two or more sensors installed on a vehicle, based on sensor values obtained from the sensors, and detects the condition of the vehicle including the sensors based on the inclination angle difference.

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

3. The state detection method according to claim 2, wherein an abnormality in any of the sensors is detected when the tilt angle difference exceeds a threshold value, and when an abnormality is detected, output of the noise cancellation signal is stopped or limited.

4. The state detection method according to claim 3, wherein the limitation of the noise cancellation signal is achieved by attenuating the signal level of the noise cancellation signal.

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

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

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

8. The condition detection method according to claim 1, wherein the sensor is attached near a wheel of the vehicle.

9. The condition detection method according to claim 1, wherein the sensor is attached to an arbitrary location on the body of the vehicle, and detects distortion of the body based on the difference in tilt angle.

10. The condition detection method according to claim 9, wherein the arbitrary location on the vehicle body includes any one of the roof, door, and window of the vehicle.

11. The condition detection method according to claim 9, wherein the sensor includes an acceleration sensor or an inclination sensor.

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 angle difference according to the state of the sensor or the vehicle.

13. The state detection method according to claim 12, wherein the calculated tilt angle difference is used as an input and the state of the vehicle including the sensor is detected using the estimation model.

14. A condition detection system comprising: an inclination angle difference calculation unit that calculates an inclination angle difference, which is the difference between the inclination angles of two or more sensors installed on a vehicle, based on sensor values obtained from the sensors; and a condition detection unit that detects the condition of the vehicle including the sensors based on the inclination angle difference.

15. A program for causing a computer to execute a process of calculating an inclination angle difference, which is the difference between the inclination angles of two or more sensors installed on a vehicle, based on sensor values obtained from the sensors, and detecting the state of the vehicle including the sensors based on the inclination angle difference.

Images