Sensor failure detection device and sensor failure detection method

Abstract

To provide a sensor failure determining device and a sensor failure determining method which can determine a failure of a sensor provided in a tire.SOLUTION: A sensor failure determining device 40 comprises a data obtaining part 41, an operation processing part 42, and a determining part 43. The data obtaining part 41 obtains data on physical quantities measured by a sensor 20 mounted on a tire 10. The operation processing part 42 has an encoding part that generates data on feature quantities using convolution-operation with respect to the obtained data by the data obtaining part 41, and a decoding part that regenerates data using inverse-operation with respect to the data on feature quantities. The determining part 43 determines whether the sensor 20 has failed or not, on the basis of the obtained data by the data obtaining part 41 and the regenerated data by the operation processing part 42.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] The present invention relates to a sensor failure determination device and a sensor failure determination method.

Background Art

[0002] Recently, research has been conducted on a system that inputs information measured in tires, vehicles, etc. into a learning-based arithmetic model and estimates tire physical information such as tire force.

[0003] Patent Document 1 describes a conventional tire physical information estimation system. The tire physical information estimation system includes a physical information estimation unit and a data acquisition unit. The physical information estimation unit has a learning-based arithmetic model from an input layer to an output layer for estimating physical information related to the tire generated by the movement of the tire. The data acquisition unit acquires input data to the input layer. The arithmetic model has a feature extraction unit that executes a convolution operation in an intermediate operation from the input layer to the output layer to extract a feature amount.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the tire physical information estimation system described in Patent Document 1, tire physical information is estimated based on a physical quantity measured by a sensor provided on the tire. The inventor considered that it is necessary to determine whether the sensor is malfunctioning because if the sensor malfunctions, the tire physical information estimation system will continue to estimate incorrect tire physical information.

[0006] This invention has been made in view of the above circumstances, and its object is to provide a sensor failure detection device and a sensor failure detection method that can determine a failure of a sensor installed on a tire. [Means for solving the problem]

[0007] One aspect of the present invention is a sensor failure detection device. The sensor failure detection device comprises a data acquisition unit that acquires data of a physical quantity measured by a sensor attached to a tire; an encoding unit that generates feature data using a convolution operation on the data acquired by the data acquisition unit; and a decoding unit that reconstructs the data using an inverse operation on the feature data; and a determination unit that determines whether or not the sensor is malfunctioning based on the data acquired by the data acquisition unit and the data reconstructed by the calculation processing unit.

[0008] Another aspect of the present invention is a sensor failure determination method. The sensor failure determination method comprises: a data acquisition step of acquiring data of a physical quantity measured by a sensor attached to a tire; an encoding process of generating feature data using a convolution operation on the data acquired in the data acquisition step; and a decoding process of reconstructing the data using an inverse operation on the feature data; and a determination step of determining whether or not the sensor is malfunctioning based on the data acquired in the data acquisition step and the data reconstructed in the calculation process step. [Effects of the Invention]

[0009] According to the present invention, it is possible to determine if a sensor installed on a tire has malfunctioned. [Brief explanation of the drawing]

[0010] [Figure 1] This is a schematic diagram illustrating the outline of a tire physical information estimation system including a sensor fault detection device according to an embodiment. [Figure 2]This is a block diagram showing the functional configuration of a tire physical information estimation device. [Figure 3] This is a schematic diagram showing the configuration of the computational model. [Figure 4] This is a block diagram showing the functional configuration of a sensor failure detection device. [Figure 5] This is a schematic diagram showing the configuration of the computational model. [Figure 6] This flowchart shows the procedure for determining sensor failures using a sensor failure detection device. [Figure 7] This graph shows the acceleration input data measured by the sensor during normal operation. [Figure 8] This graph shows the reconstructed data calculated by the computational model for the input data in Figure 7. [Figure 9] This graph shows the input data for acceleration measured by the malfunctioning sensor 20. [Figure 10] This graph shows the reconstructed data calculated by the computational model for the input data in Figure 9. [Modes for carrying out the invention]

[0011] The present invention will be described below with reference to Figures 1 to 10, based on preferred embodiments. The same or equivalent components and members shown in each drawing will be denoted by the same reference numerals, and redundant explanations will be omitted as appropriate. Furthermore, the dimensions of the members in each drawing will be enlarged or reduced as appropriate for ease of understanding. Additionally, some members that are not important for explaining the embodiments will be omitted from the drawings.

[0012] (Embodiment) FIG. 1 is a schematic diagram for explaining the outline of a tire physical information estimation system 100 including a sensor failure determination device 40 according to an embodiment. The tire physical information estimation system 100 includes a sensor 20 disposed on a tire 10, a tire physical information estimation device 30, and a sensor failure determination device 40. Further, the tire physical information estimation system 100 may include a server device 80 or the like for acquiring and storing tire physical information such as a tire force F estimated by the tire physical information estimation device 30 and moments about three axes acting on the tire 10 via a communication network 91 and monitoring the tire physical information.

[0013] The sensor 20 measures physical quantities of the tire 10 such as acceleration, strain, tire air pressure, and tire temperature in the tire 10, and outputs the measured data to the tire physical information estimation device 30 and the sensor failure determination device 40.

[0014] The tire physical information estimation device 30 estimates tire physical information based on the data measured by the sensor 20. The tire physical information estimation device 30 uses the data measured by the sensor 20 in the calculation for estimating the tire physical information, but may also acquire information from the vehicle side such as vehicle acceleration from a vehicle control device 90 or the like and use it in the calculation for estimating the tire physical information.

[0015] The tire physical information estimation device 30 outputs the estimated tire physical information such as the estimated tire force F and moments about three axes acting on the tire 10 to, for example, the vehicle control device 90. The vehicle control device 90 uses the tire physical information input from the tire physical information estimation device 30 for, for example, estimating a braking distance, applying it to vehicle control, and further notifying a driver of information regarding safe driving of the vehicle. The vehicle control device 90 can also provide information regarding safe driving of the vehicle in the future using map information, weather information, or the like. Further, when the vehicle control device 90 of the tire physical information estimation system 100 has a function of automatically driving the vehicle, the estimated tire physical information is provided to the vehicle control device 90 as data used for vehicle speed control or the like in automatic driving.

[0016] The sensor failure determination device 40 determines whether the sensor 20 is faulty based on the data measured by the sensor 20, and notifies the determination result to the tire physical information estimation device 30, the server device 80, etc. The sensor failure determination device 40 determines whether the sensor 20 is faulty by means of an arithmetic model that performs encoding and decoding using a convolution operation on the data measured by the sensor 20.

[0017] FIG. 2 is a block diagram showing the functional configuration of the tire physical information estimation device 30. The sensor 20 includes an acceleration sensor 21, a strain gauge 22, a pressure gauge 23, a temperature sensor 24, etc., and measures physical quantities in the tire 10. These sensors measure physical quantities related to the deformation and movement of the tire 10 as physical quantities of the tire 10.

[0018] The acceleration sensor 21 and the strain gauge 22 measure the acceleration and the amount of strain generated in the tire 10, respectively, while mechanically moving together with the tire 10. The acceleration sensor 21 is disposed, for example, on the tread, side, bead, wheel, etc. of the tire 10, and measures the acceleration in the three axes of the circumferential direction, axial direction, and radial direction of the tire 10.

[0019] The strain gauge 22 is disposed on the tread, side, bead, etc. of the tire 10, and measures the strain at the disposed location. Also, the pressure gauge 23 and the temperature sensor 24 are disposed, for example, on the air valve of the tire 10, and measure the tire air pressure and the tire temperature, respectively. The temperature sensor 24 may be disposed directly on the tire 10 in order to accurately measure the temperature of the tire 10. The tire 10 may be attached with, for example, an RFID 11 to which unique identification information is given in order to identify each tire.

[0020] The tire physical information estimation device 30 includes a data acquisition unit 31, a physical information estimation unit 32, and a communication unit 33. The tire physical information estimation device 30 is an information processing device such as a PC (personal computer). Each part of the tire physical information estimation device 30 can be realized in hardware terms using electronic elements and mechanical parts, including the CPU of a computer, and in software terms using computer programs, etc. However, here we are describing a functional block realized through the cooperation of these components. Therefore, it will be understood by those skilled in the art that these functional blocks can be realized in various forms by combining hardware and software.

[0021] The data acquisition unit 31 acquires information on acceleration, strain, air pressure, and temperature measured by the sensor 20 via wireless communication or the like. The communication unit 33 communicates with external devices such as the vehicle control device 90 and the server device 80 via wired or wireless communication. The communication unit 33 transmits the physical quantities of the tire 10 measured by the sensor 20, and estimated tire physical information about the tire 10, to the external devices via a communication line, such as CAN (Control Area Network) or the Internet.

[0022] The physical information estimation unit 32 has a calculation model 32a, and inputs information from the data acquisition unit 31 to the calculation model 32a to estimate tire physical information such as tire force F and moments around the three axes acting on the tire 10. As shown in Figure 2, the tire force F has three axial components: longitudinal force Fx in the longitudinal direction of the tire 10, lateral force Fy in the lateral direction, and vertical load Fz. The physical information estimation unit 32 may calculate all of these three axial components, or it may calculate at least one component or two components in any combination.

[0023] The computational model 32a uses a learning model such as a neural network. Figure 3 is a schematic diagram showing the configuration of computational model 32a. Computational model 32a is a CNN (Convolutional Neural Network) type, and is a learning model that incorporates convolution and pooling operations used in its prototype, the so-called LeNet. Figure 3 shows an example in which acceleration data in three axes is used as input data to computational model 32a, and tire forces in three axes are output.

[0024] The computation model 32a comprises an input layer 50, a feature extraction unit 51, an intermediate layer 52, a fully connected unit 53, and an output layer 54. The input layer 50 receives time-series data of acceleration in the three axes acquired by the data acquisition unit 31. The acceleration data is measured time-series by the sensor 20, and data from a certain time interval is extracted using a window function to be used as input data.

[0025] The acceleration measured by tire 10 has periodicity with each rotation of tire 10. The time interval of the input data extracted by the window function should be, for example, the time corresponding to the rotation period of tire 10, so that the input data itself has periodicity. The window function may also be used to extract input data in a time interval shorter or longer than one rotation of tire 10, and as long as the extracted input data contains periodic information, the computational model 32a can be trained.

[0026] The feature extraction unit 51 extracts features using convolution and pooling operations, etc., and transmits them to each node of the hidden layer 52. The feature extraction unit 51 performs convolution operations on the input data using multiple filters. The convolution operation is performed on time-series input data such as acceleration data, while moving the filters. The pooling operation is performed on the data after the convolution operation, for example, by performing a maximum value pooling operation that selects the larger of two values ​​arranged in time series. The feature extraction unit 51 repeatedly performs convolution and pooling operations, etc., to extract features.

[0027] The fully connected unit 53 fully connects the data from each node of the intermediate layer 52 in one or more layers and outputs the tire forces Fx, Fy, and Fz to each node of the output layer 54. The fully connected unit 53 performs calculations using fully connected paths that perform linear calculations using weighting, but in addition to linear calculations, it may also perform nonlinear calculations using activation functions, etc.

[0028] Each node in the output layer 54 may output tire physical information such as the tire force in the three axes, as well as the moment acting on the tire 10 around the three axes. The output layer 54 may output one type or any combination of multiple types of tire physical information from the tire force in the three axes and the moment acting on the tire 10 around the three axes.

[0029] The computational model 32a can be trained by mounting tires 10 with specifications appropriate for that vehicle onto an actual vehicle and conducting a test run of the vehicle. The specifications of the tire 10 include information about the tire's performance, such as tire size, tire width, aspect ratio, tire strength, tire outer diameter, load index, and manufacturing date.

[0030] The server device 80 acquires from the tire physical information estimation device 30 physical quantities of the tire 10 measured by the sensor 20, as well as tire physical information such as the estimated tire force F and the moment acting on the tire 10 around the three axes. The server device 80 may also be configured to store physical quantities measured by the tire 10 and tire physical information estimated by the tire physical information estimation device 30 from multiple vehicles.

[0031] Figure 4 is a block diagram showing the functional configuration of the sensor failure detection device 40. As described above, the sensor 20 includes an acceleration sensor 21, a strain gauge 22, a pressure gauge 23, and a temperature sensor 24, etc., and measures physical quantities in the tire 10.

[0032] The sensor failure detection device 40 includes a data acquisition unit 41, an arithmetic processing unit 42, a determination unit 43, and a communication unit 44. The sensor failure detection device 40 is an information processing device such as a PC (personal computer). Each part of the sensor failure detection device 40 can be realized in hardware terms using electronic elements and mechanical parts, including the CPU of a computer, and in software terms using computer programs, etc. However, here we are describing a functional block realized through the cooperation of these components. Therefore, it will be understood by those skilled in the art that these functional blocks can be realized in various forms by combining hardware and software.

[0033] The data acquisition unit 41 acquires acceleration, strain, air pressure, and temperature data measured by the sensor 20 via wireless communication or the like, and outputs it to the preprocessing unit 42a of the calculation processing unit 42. The communication unit 44 communicates with external devices such as the tire physical information estimation device 30 and the server device 80 via wired or wireless communication. The communication unit 44 transmits the determination result of whether or not the sensor 20 is malfunctioning to the external device.

[0034] The arithmetic processing unit 42 includes a preprocessing unit 42a and an arithmetic model 42b. The preprocessing unit 42a normalizes the data input from the data acquisition unit 41 and outputs it to the arithmetic model 42b. The normalization in the preprocessing unit 42a will be explained using the example of input data being acceleration data Tx, Ty, and Tz in the three axes. Note that the acceleration data Tx, Ty, and Tz are the circumferential, axial (vehicle axis direction) and radial acceleration of the tire 10, respectively.

[0035] For example, if the acceleration data Tx generated during vehicle operation is between -0.2G and 0.2G, the preprocessor 42a divides the acceleration data Tx by a constant value of 0.2 and outputs the normalized acceleration data Txn. If the acceleration data Ty generated during vehicle operation is between -0.1G and 0.1G, the preprocessor 42a divides the acceleration data Ty by a constant value of 0.1 and outputs the normalized acceleration data Tyn.

[0036] Furthermore, if the acceleration data Tz generated during vehicle operation is between -0.8G and 1G, the preprocessor 42a subtracts the acceleration data Tz by the median value of 0.1, divides it by a constant value of 0.9, and outputs the normalized acceleration data Tzn. Through these processes, the preprocessor 42a outputs the normalized acceleration data Txn, Tyn, and Tzn as values ​​in the range of -1 and 1.

[0037] The normalization of physical quantities measured by the tire 10, such as acceleration and strain in the three axes, is not limited to the example described above, and can be set appropriately depending on the nature of the physical quantities measured by the tire 10 and the range of possible values.

[0038] The preprocessor 42a does not need to perform normalization processing if, for example, only acceleration data for one of the three axes is used as input data for the calculation model 42b. The preprocessor 42a performs normalization processing when physical quantities are measured in at least two of the three axes of the tire 10 and input to the calculation model 42b.

[0039] The computational model 42b uses an autoencoder-type learning model that employs convolution operations, etc. Figure 5 is a schematic diagram showing the configuration of computational model 42b. For example, computational model 42b receives normalized acceleration data in the three axes as input, generates feature data using convolution operations, and reconstructs the input data by inverse operations.

[0040] The computation model 42b comprises an input layer 60, an encoding unit 61, a decoding unit 62, and an output layer 63. The input layer 60 receives normalized time-series data of acceleration in the three axes output from the preprocessing unit 42a. The acceleration data is measured time-series by the sensor 20, and data from a certain time interval is extracted using a window function to be used as input data.

[0041] The acceleration measured by tire 10 has periodicity with each rotation of tire 10. The time interval of the input data extracted by the window function should be, for example, the time corresponding to the rotation period of tire 10, so that the input data itself has periodicity. The window function may also be used to extract input data in a time interval shorter or longer than one rotation of tire 10, as long as the extracted input data contains periodic information.

[0042] The encoding unit 61 generates feature data using convolution and pooling operations. The encoding unit 61 performs a convolution operation on the input data using multiple filters. The convolution operation is performed on time-series input data such as acceleration data, while moving the filters. The pooling operation is performed on the data after the convolution operation, for example, by performing a maximum value pooling operation to select the larger of two time-series values. The encoding unit 61 repeatedly performs convolution and pooling operations to generate feature data.

[0043] The decoding unit 62 uses the inverse operation of the encoding unit 61 to reconstruct the feature data generated by the encoding unit 61 and outputs it to the output layer 63. The inverse operation performed by the decoding unit 62 is called transpose convolution or deconvolution.

[0044] In the encoding unit 61, the input data is downsampled to generate feature data, while the decoding unit 62 performs the reverse operation, upsampling the feature data to reconstruct the input data. The decoding unit 62 may also use a method called up convolution as an operation equivalent to transposed convolution.

[0045] The computational model 42b can perform so-called unsupervised learning by being trained so that the data input to the input layer 60 and the data output from the output layer 63 are approximate. The computational model 42b is trained while the sensor 20, such as the acceleration sensor 21, is operating normally. If the sensor 20 malfunctions due to excessive vibration or other reasons, the data input to the input layer 60 will be different from the data under normal operation conditions, and the trained computational model 42b will reconstruct and output data with increased errors relative to the input data.

[0046] Returning to Figure 4, the determination unit 43 determines whether the sensor 20 is malfunctioning based on the input data to the input layer 60 of the calculation model 42b and the regenerated data output from the output layer 63. The determination unit 43 determines that the sensor 20 is operating normally if the sum of the absolute values ​​of the errors between the input data to the input layer 60 and the regenerated data output from the output layer 63 is less than a predetermined threshold. The determination unit 43 determines that the sensor 20 is malfunctioning if the sum of the absolute values ​​of the errors between the input data to the input layer 60 and the regenerated data output from the output layer 63 is greater than or equal to a predetermined threshold.

[0047] For example, if the input data is acceleration data for the three axes of the tire 10, the determination unit 43 may calculate the sum of the absolute values ​​of the errors between the input data to the input layer 60 and the regenerated data output from the output layer 63 for each axis and make a determination for each axis. Alternatively, the determination unit 43 may calculate the sum of the absolute values ​​of the errors between the input data to the input layer 60 and the regenerated data output from the output layer 63 for each axis, and then add up all the calculated sums for each axis to make a single overall determination. Similarly, if the input data is acceleration data for two of the three axes of the tire 10, the determination unit 43 may make a determination for each axis or make a single overall determination.

[0048] The server device 80 may acquire and store the determination result of whether or not the sensor 20 is faulty from the sensor fault determination device 40. The tire physical information estimation device 30 also acquires the determination result of whether or not the sensor 20 is faulty, and if the sensor 20 is faulty, it stops the tire physical information estimation process or switches to an alternative estimation process if one exists. The tire physical information estimation device 30 may output the fault of the sensor 20 to the vehicle control device 90, and the vehicle control device 90 may notify the driver of the fault of the sensor 20.

[0049] Next, the operation of the sensor failure detection device 40 will be explained. Figure 6 is a flowchart showing the procedure for determining sensor failure by the sensor failure detection device 40. The sensor failure detection device 40 acquires data on physical quantities such as acceleration, strain, tire pressure, and tire temperature measured by the sensor 20 in the tire 10 using the data acquisition unit 41 (S1).

[0050] The arithmetic processing unit 42 performs the process of extracting input data for a certain time interval from the data acquired by the data acquisition unit 41 (S2). The preprocessing unit 42a performs normalization processing on the input data (S3). The normalized input data is input to the input layer 60 of the arithmetic model 42b.

[0051] The encoding unit 61 of the computation model 42b performs a process to generate feature data by performing convolution and pooling operations on the input data (S4). The decoding unit 62 of the computation model 42b performs the inverse encoding operation on the feature data generated by the encoding unit 61 to reconstruct the data (S5).

[0052] The determination unit 43 determines whether the sensor 20 is faulty based on the input data to the calculation model 42b and the reconstructed data output by the decoding unit 62 (S6), and then terminates the process.

[0053] Figure 7 is a graph representing the input data of acceleration measured by sensor 20 during normal operation, and Figure 8 is a graph representing the reconstructed data calculated by the computational model 42b for the input data in Figure 7. The input data shown in Figure 7 is normalized acceleration data Tx, Ty, and Tz in the three axes. In the computational model 42b, feature data is generated by a convolution operation in the coding unit 61, and the data is reconstructed by the inverse operation. Therefore, as shown in Figure 8, when the input data is normal, the waveform fluctuation characteristics in each input data are reflected in the reconstructed data.

[0054] Figure 9 is a graph representing the input data of acceleration measured by the faulty sensor 20, and Figure 10 is a graph representing the reconstructed data calculated by the computational model 42b for the input data in Figure 9. The input data shown in Figure 9 is normalized acceleration data Tx, Ty, and Tz in the three axes. For input data from the faulty sensor 20, the reconstructed data output by the computational model 42b shows waveform fluctuations that differ from the characteristics of the input data, resulting in an increased error compared to the input data.

[0055] The normal range Rc is defined as the sum of the absolute values ​​of the errors in the input data from the sensor 20 during normal operation and the reconstructed data calculated by the calculation model 42b. The abnormal range Re is defined as the sum of the absolute values ​​of the errors in the input data from the malfunctioning sensor 20 and the reconstructed data calculated by the calculation model 42b. The determination unit 43 sets a threshold D between the normal range Rc and the abnormal range Re to determine whether the sensor 20 is malfunctioning. Even if the upper limit of the normal range Rc and the lower limit of the abnormal range Re partially overlap, the threshold D can be set considering the distribution of the normal range Rc and the abnormal range Re.

[0056] The sensor failure detection device 40 encodes the input data using a convolution operation in the calculation model 42b and reconstructs the data using an inverse operation. Based on the input data and the reconstructed data, it can determine whether or not the sensor 20 is malfunctioning.

[0057] The arithmetic processing unit 42 normalizes the data acquired by the data acquisition unit 41 in the preprocessing unit 42a, thereby making the influence of the data in each axis direction on the fault determination by the determination unit 43 equal.

[0058] The sensor 20 measures physical quantities in at least two of the three axes of the tire 10, and the determination unit 43 may make a fault determination for each axis. The sensor fault determination device 40 can provide information that the sensor 20 is in a faulty state in part. The tire physical information estimation device 30 can, for example, continue estimating tire physical information based on data acquired from the normally operating sensor 20, even if the sensor 20 is in a faulty state in part.

[0059] Alternatively, the sensor 20 may measure physical quantities in at least two of the three axes of the tire 10, and the determination unit 43 may make a single determination based on the data for each axis. The sensor failure determination device 40 can determine whether the input data for each axis to the calculation model 42b is in a state suitable for feature extraction.

[0060] Next, the features of the sensor failure detection device 40 and the sensor failure detection method according to the embodiment will be described. The sensor failure detection device 40 according to this embodiment comprises a data acquisition unit 41, an arithmetic processing unit 42, and a determination unit 43. The data acquisition unit 41 acquires data of physical quantities measured by a sensor 20 attached to the tire 10. The arithmetic processing unit 42 has an encoding unit 61 that generates feature data using a convolution operation on the data acquired by the data acquisition unit 41, and a decoding unit 62 that reconstructs the data using an inverse operation on the feature data. The determination unit 43 determines whether or not the sensor 20 is faulty based on the data acquired by the data acquisition unit 41 and the data reconstructed by the arithmetic processing unit 42. As a result, the sensor failure detection device 40 can determine if the sensor 20 provided on the tire 10 is faulty.

[0061] Furthermore, the arithmetic processing unit 42 includes a preprocessing unit 42a that normalizes the data acquired by the data acquisition unit 41. This allows the sensor failure determination device 40 to ensure that the influence of data in each axial direction is equal in the failure determination by the determination unit 43.

[0062] Furthermore, the sensor 20 measures physical quantities in at least two of the three axes of the tire 10. The determination unit 43 determines whether or not the sensor 20 is malfunctioning in each axis direction. As a result, the sensor malfunction determination device 40 can provide information that the sensor 20 is malfunctioning in part.

[0063] Furthermore, the determination unit 43 comprehensively determines whether or not the sensor 20 is malfunctioning based on the data in each axial direction. This allows the sensor malfunction determination device 40 to determine whether or not the input data in each axial direction to the calculation model 42b is in a state suitable for feature extraction.

[0064] The sensor failure detection method comprises a data acquisition step, a calculation processing step, and a determination step. The data acquisition step acquires data of physical quantities measured by a sensor 20 attached to the tire 10. The calculation processing step performs an encoding process to generate feature data using a convolution operation on the data acquired in the data acquisition step, and a decoding process to reconstruct the data using an inverse operation on the feature data. The determination step determines whether the sensor 20 is faulty based on the data acquired in the data acquisition step and the data reconstructed in the calculation processing step. This sensor failure detection method makes it possible to determine if the sensor 20 installed on the tire 10 is faulty.

[0065] The embodiments of the present invention have been described above. These embodiments are illustrative, and it will be understood by those skilled in the art that various modifications and changes are possible within the scope of the claims of the present invention, and that such modifications and changes are also within the scope of the claims of the present invention. Accordingly, the descriptions and drawings herein should be treated as illustrative rather than limiting. [Explanation of Symbols]

[0066] 10 tires, 20 sensors, 40 sensor fault detection device, 41 Data acquisition unit, 42 Calculation processing unit, 42a Preprocessing unit, 42b Calculation model, 43 Determination unit, 61 Encoding unit, 62 Decoding unit.

Claims

1. A data acquisition unit that acquires data on physical quantities measured by sensors attached to the tires, An arithmetic processing unit having an encoding unit that generates feature data using a convolution operation on the data acquired by the data acquisition unit, and a decoding unit that reconstructs the data using an inverse operation on the feature data, A determination unit determines whether or not the sensor is malfunctioning based on the data acquired by the data acquisition unit and the data reconstructed by the calculation processing unit. A sensor failure detection device characterized by comprising the following features.

2. The sensor failure determination device according to claim 1, characterized in that the calculation processing unit includes a preprocessing unit that normalizes the data acquired by the data acquisition unit.

3. The aforementioned sensor measures physical quantities in at least two of the three axial directions of the tire. The sensor failure determination device according to claim 1 or 2, characterized in that the determination unit makes determinations for each axial direction.

4. The aforementioned sensor measures physical quantities in at least two of the three axial directions of the tire. The sensor failure determination device according to claim 1 or 2, characterized in that the determination unit makes a comprehensive determination based on the data in each axial direction.

5. A data acquisition step involves acquiring data on physical quantities measured by sensors attached to the tires, The computational processing steps include: an encoding process that generates feature data using a convolution operation on the data acquired in the data acquisition step, and a decoding process that reconstructs the data using an inverse operation on the feature data; A determination step that determines whether or not the sensor is malfunctioning based on the data acquired in the data acquisition step and the data reconstructed in the calculation processing step, A sensor failure detection method characterized by comprising the following:

Images