Tire physical information estimation system and computational model generation system

Abstract

To provide a tire physical information estimation system that can estimate physical information concerning a tire with good accuracy, and a calculation model generation system.SOLUTION: A tire physical information estimation system 100 comprises a physical information estimation part 32 and a data obtaining part 31. The physical information estimation part 32 has a learning-type calculation model 32a extending from an input layer to an output layer, which estimates tire physical information that is generated by movements of a tire 10. The data obtaining part 31 obtains input data on the input layer. The calculation model 32a has a feature extraction part that executes convolution calculation in calculation in the course from the input layer to the output layer, which outputs tire physical information in at least two axial directions normalized from the output layer.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a tire physical information estimation system and a computational model generation system. [Background technology]

[0002] Recently, research has been conducted into systems that input information measured on tires, vehicles, etc. into a learning-type calculation model to estimate 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-type computational model extending from an input layer to an output layer to estimate physical information related to a tire that is generated by tire motion. The data acquisition unit acquires input data for the input layer. The computational model has a feature extraction unit that extracts feature quantities by performing convolution computations during intermediate computations from the input layer to the output layer. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-46080 Summary of the Invention [Problem to be solved by the invention]

[0005] When a system with a small calculation scale is constructed using the tire physical information estimation system described in Patent Document 1, even if real-time estimation is ensured, there is a possibility that the estimation accuracy will deteriorate. The inventors of the present application have recognized that there is room for improvement in the technology disclosed in Patent Document 1 in terms of increasing the estimation accuracy of tire physical information based on a learning-type calculation model.

[0006] The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a tire physical information estimation system and a calculation model generation system that can accurately estimate physical information about a tire. [Means for solving the problem]

[0007] One aspect of the present invention is a tire physical information estimation system, which includes a learning-type computational model extending from an input layer to an output layer, a physical information estimation unit that estimates tire physical information generated by tire motion, and a data acquisition unit that acquires input data to the input layer, the computational model having a feature extraction unit that executes convolution operations in intermediate operations from the input layer to the output layer, and outputs normalized tire physical information in at least two axial directions from the output layer. [Effects of the Invention]

[0008] According to the present invention, physical information about a tire can be estimated with high accuracy. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a schematic diagram for explaining an overview of a tire physical information estimation system according to an embodiment; [Figure 2] 1 is a block diagram showing a functional configuration of a tire physical information estimation system according to an embodiment. [Figure 3] FIG. 2 is a schematic diagram showing the configuration of a calculation model. [Figure 4] FIG. 1 is a block diagram showing a functional configuration of a computation model generation system. [Figure 5] 4 is a flowchart showing a procedure of a tire physical information estimation process performed by the tire physical information estimation device. [Figure 6] 6(a) to 6(c) are graphs showing the correlation between the estimated values ​​by the tire physical information estimation system and the measured values. [Figure 7]7(a) to 7(c) are graphs showing the correlation between estimated values ​​and measured values ​​when a learning model of a comparative example is used. [Figure 8] 1 is a chart showing the mean absolute error of the estimates and measurements. [Figure 9] FIG. 10 is a block diagram showing a functional configuration of a tire physical information estimation system according to a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0010] The present invention will be described below based on preferred embodiments with reference to Figures 1 to 9. The same or equivalent components and members shown in each drawing are designated by the same reference numerals, and duplicate descriptions will be omitted where appropriate. Furthermore, the dimensions of the members in each drawing are enlarged or reduced as appropriate for ease of understanding. Furthermore, some members that are not important for explaining the embodiments will be omitted from the drawings.

[0011] (Embodiment) 1 is a schematic diagram illustrating an overview of a tire physical information estimation system 100 according to an embodiment. The tire physical information estimation system 100 includes a sensor 20 disposed on a tire 10 and a tire physical information estimation device 30. The tire physical information estimation system 100 may also include a server device 40 that acquires and stores tire physical information, such as tire force F, road friction coefficient, and moments acting on the tire 10 about three axes, estimated by the tire physical information estimation device 30 via a communication network 91 and monitors the tire physical information.

[0012] The sensor 20 measures physical quantities of the tire 10, such as the acceleration and strain of the tire 10, the tire air pressure, and the tire temperature, and outputs the measured data to the tire physical information estimation device 30. 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 calculations to estimate the tire physical information, but may also obtain information from the vehicle side, such as vehicle acceleration, from a vehicle control device 90 or the like and use the information in calculations to estimate the tire physical information.

[0013] The tire physical information estimation device 30 outputs tire physical information such as the estimated tire force F, road friction coefficient, and moments acting on the tire 10 about three axes to, for example, a vehicle control device 90. The vehicle control device 90 uses the tire physical information input from the tire physical information estimation device 30, for example, to estimate braking distance, apply it to vehicle control, and further notify the driver of information related to safe vehicle driving. The vehicle control device 90 can also provide information related to safe vehicle driving in the future using map information, weather information, etc. Furthermore, when the vehicle control device 90 has a function for automatically driving the vehicle, the tire physical information estimation system 100 provides the estimated tire physical information to the vehicle control device 90 as data to be used for vehicle speed control and the like during automatic driving.

[0014] 2 is a block diagram showing the functional configuration of a tire physical information estimation system 100 according to an embodiment. The sensor 20 of the tire physical information estimation system 100 includes an acceleration sensor 21, a strain gauge 22, a pressure gauge 23, a temperature sensor 24, etc., and measures physical quantities of the tire 10. These sensors measure physical quantities related to the deformation and movement of the tire 10 as the physical quantities of the tire 10.

[0015] The acceleration sensor 21 and the strain gauge 22 move mechanically together with the tire 10, and measure the acceleration and strain amount, respectively, occurring in the tire 10. The acceleration sensor 21 is disposed, for example, in the tread, side, bead, wheel, etc. of the tire 10, and measures the acceleration in three axes of the tire 10: the circumferential direction, the axial direction, and the radial direction.

[0016] The strain gauges 22 are disposed in the tread, sides, beads, etc. of the tire 10 and measure strain at the locations where they are disposed. The pressure gauge 23 and temperature sensor 24 are disposed, for example, in the air valve of the tire 10 and measure the tire pressure and tire temperature, respectively. The temperature sensor 24 may be disposed directly on the tire 10 to accurately measure the temperature of the tire 10. The tires 10 may be fitted with, for example, an RFID 11 or the like that has unique identification information attached thereto in order to identify each tire.

[0017] 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 unit in the tire physical information estimation device 30 can be realized in terms of hardware using electronic elements and mechanical parts such as a computer CPU, and in terms of software using a computer program, but the functional blocks realized by the cooperation of these elements are depicted here. 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.

[0018] 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 40 via wired or wireless communication or the like. The communication unit 33 transmits the physical quantities of the tire 10 measured by the sensor 20, tire physical information estimated about the tire 10, and the like to the external devices via a communication line, for example, a CAN (Control Area Network), the Internet, or the like.

[0019] 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, road friction coefficient, and moments about three axes acting on the tire 10. As shown in FIG. 2, the tire force F has three axial components: a longitudinal force Fx in the longitudinal direction of the tire 10, a lateral force Fy in the lateral direction, and a load Fz in the vertical direction. The physical information estimation unit 32 may calculate all of these three axial components, or may calculate at least one of the components or two components in any combination.

[0020] The computational model 32a uses a learning model such as a neural network. Fig. 3 is a schematic diagram showing the configuration of the computational model 32a. The computational model 32a is a CNN (Convolutional Neural Network) type, and is a learning model that includes convolution and pooling operations used in its prototype, so-called LeNet. Fig. 3 shows an example in which acceleration data in three axial directions is used as input data to the computational model 32a, and normalized tire forces in three axial directions are output.

[0021] The computational model 32a includes an input layer 50, a feature extraction unit 51, a middle layer 52, a full connection unit 53, and an output layer 54. The computational model 32a is a multitasking model in which the feature extraction unit 51 is common to the three axial directions and a full connection unit 53 is provided for each of the three axial directions, and normalized values ​​of the tire forces Fx, Fy, and Fz in the three axial directions are output from the output layer 54. The normalized tire forces in the three axial directions output from the output layer 54 are denoted as Fxn, Fyn, and Fzn.

[0022] For example, when the tire forces Fx and Fy generated when the vehicle is traveling are equal to or greater than -5000 N and equal to or less than 5000 N, the normalized tire forces Fxn and Fyn output from the output layer 54 are values ​​obtained by dividing the tire forces Fx and Fy by a constant value of 5000. As a result, the normalized tire forces Fxn and Fyn vary within a range of equal to or greater than -1 and equal to or less than 1.

[0023] For example, if the tire force Fz generated when the vehicle is traveling is 2000 N or more and 8000 N or less, the normalized tire force Fzn output from the output layer 54 is a value obtained by subtracting the median value 5000 from the tire force Fz and dividing the result by a constant value 3000. As a result, the normalized tire force Fzn varies within a range of -1 or more and 1 or less.

[0024] The normalization of tire physical information such as the tire force F in the three axial directions, the road friction coefficient, and the moment acting on the tire 10 about three axes is not limited to the above example, and can be set appropriately depending on the properties of the tire physical information, the range of possible values, etc.

[0025] The input layer 50 receives time-series data of acceleration in three axial directions acquired by the data acquisition unit 31. The acceleration data is measured in time series by the sensor 20, and data within a certain time interval is extracted using a window function to be used as input data. The input data may be, for example, 224 pieces of acceleration data included in the certain time interval in each axial direction.

[0026] The acceleration measured at the tire 10 is periodic for each rotation of the tire 10. The time interval of the input data extracted by the window function may be, for example, a time period corresponding to the rotation period of the tire 10, so that the input data itself has periodicity. Note that the window function may extract input data in a time interval shorter or longer than one rotation of the tire 10, and as long as the extracted input data contains at least periodic information, learning of the calculation model 32a is possible.

[0027] The feature extraction unit 51 extracts features using a convolution operation 51a and a pooling operation 51b and transmits them to each node in the intermediate layer 52. In the example of the feature extraction unit 51 shown in FIG. 3, a first convolution operation is performed on input data using 20 filters. The convolution operation 51a performs a convolution operation while moving the filter on time-series input data such as acceleration data. The convolution operation 51a is performed on each of the acceleration data in three axial directions, which are multiple input data, but the scale of the operation can be reduced by using the same filter for each axial direction.

[0028] In the convolution operation 51a, the filter length is set to 5, but it may be set to a value of approximately 1 to 5 as appropriate. Note that the convolution operation multiplies consecutive data (e.g., A1, A2, A3) of the filter length in the time-series input data by each value in the filter (f1, f2, f3), and then adds the values ​​obtained by multiplication to obtain A1×f1+A2×f2+A3×f3. Note that the convolution operation may be performed by performing zero padding, which adds "0 (zero)" data to the end of the input data. Also, the amount of filter movement in the convolution operation is usually set to one input data, but this can be changed as appropriate to reduce the calculation model 32a.

[0029] The pooling operation 51b executes a first maximum value pooling operation on the data after the first convolution operation. The pooling operation 51b selects the larger value of two values ​​arranged in time series, for example.

[0030] The second convolution operation 51c performs a convolution operation on the data after the pooling operation 51b using, for example, 50 filters. The filter length in the convolution operation 51c may be the same as or different from that in the convolution operation 51a. For the convolution operation 51c, the same filter is used in each axis direction, making it possible to reduce the scale of the operation.

[0031] The pooling operation 51d performs a second maximum value pooling operation on the data after the convolution operation 51c. Similar to the pooling operation 51b, the pooling operation 51d selects the larger of two values ​​arranged in time series, for example. Through the convolution operation and pooling operation, the feature extraction unit 51 obtains 64 data in each axis direction, i.e., 64 × 3ch data, as the operation result and outputs it to each node in the hidden layer 52.

[0032] The full connection unit 53 fully connects the data from each node in the intermediate layer 52 for each of the three axis directions in two layers, and outputs normalized tire forces Fxn, Fyn, and Fzn to each node in the output layer 54. The full connection unit 53 performs calculations using a full connection path that executes linear calculations using weighting, etc., but may also execute nonlinear calculations using an activation function, etc., in addition to the linear calculations.

[0033] The physical information estimation unit 32 can restore and estimate the tire forces Fx, Fy, and Fz by performing an inverse operation of normalization on the normalized tire forces Fxn, Fyn, and Fzn output to the output layer 54.

[0034] Each node of the output layer 54 may output tire physical information such as the road friction coefficient and the moments about three axes acting on the tire 10 in addition to the normalized tire forces in the three axial directions. The tire physical information such as the road friction coefficient and the moments about three axes acting on the tire 10 output to the output layer 54 may be normalized values. The output layer 54 may output one type of tire physical information or multiple types of tire physical information in any combination out of the tire physical information such as the tire forces in the three axial directions, the road friction coefficient, and the moments about three axes acting on the tire 10.

[0035] In addition, when estimating the road surface friction coefficient, the output layer 54 may output an estimated value of the road surface friction coefficient, or may classify the road surface friction coefficient into categories such as dry, wet, snowy, or icy conditions and output which category it falls into.

[0036] Returning to Figure 2, the server device 40 acquires, from the tire physical information estimation device 30, the physical quantities of the tire 10 measured by the sensor 20, as well as tire physical information such as the tire force F and road friction coefficient estimated for the tire 10. The server device 40 may accumulate the physical quantities measured for the tire 10 and the tire physical information estimated by the tire physical information estimation device 30 from a plurality of vehicles.

[0037] 4 is a block diagram showing the functional configuration of the computational model generation system 110. The computational model generation system 110 includes a tire physical information measurement device 60 and a computational model generation device 70 having a learning processing unit 71. The computational model generation device 70 has the learning processing unit 71 in addition to the components of the tire physical information estimation device 30. The components of the computational model generation device 70 that correspond to the components of the tire physical information estimation device 30 have the same functions as those of the tire physical information estimation device 30, but the computational model 32a is either pre-learning or currently being learned.

[0038] The tire physical information measurement device 60 measures tire physical information such as the tire force F in three axial directions, the road friction coefficient, and the moment acting on the tire 10 about three axes. The learning processing unit 71 normalizes the tire physical information measured by the tire physical information measurement device 60 and uses the normalized information as training data to train the calculation model 32a. In the learning process of the calculation model 32a, the tire physical information is estimated by the calculation model 32a based on input information and compared with the training data.

[0039] The learning processing unit 71 compares the tire physical information estimated by the computation model 32a with the teacher data, newly sets various coefficients in the computation process, such as weighting, to the computation model 32a, and repeatedly updates the model to perform learning. The tire physical information estimation system 100 estimates tire physical information using the computation model 32a that has been trained by the computation model generation system 110.

[0040] The computation model 32a may basically be changed in configuration and weighting, such as the number of layers in all joints 53, depending on the specifications of the tire 10. The computation model 32a can be trained in a rotation test using tires 10 (including wheels) of various specifications. However, it is not necessary to train the computation model 32a strictly for each specification of the tire 10. For example, by building a computation model 32a through training for each type, such as passenger car tires or truck tires, and estimating the tire force F within a certain error range, one computation model 32a may be shared for tires 10 of multiple specifications, thereby reducing the number of computation models.

[0041] The calculation model 32a can also be trained by mounting the tire 10 on an actual vehicle and running the vehicle on a test run. The specifications of the tire 10 include information on tire performance, such as tire size, tire width, aspect ratio, tire strength, tire outer diameter, load index, and manufacturing date.

[0042] The computational model 32a may be trained by performing a rotation test while changing the road friction coefficient of the contact surface that contacts the tire 10. Furthermore, the computational model 32a may be trained by mounting the tire 10 on an actual vehicle and running the vehicle on test roads with different road friction coefficients.

[0043] Next, a description will be given of the operation of the tire physical information estimation system 100. Fig. 5 is a flowchart showing the procedure of tire physical information estimation processing by the tire physical information estimation device 30. The tire physical information estimation device 30 acquires physical quantities such as acceleration, strain, tire air pressure, and tire temperature of the tire 10 measured by the sensor 20 using the data acquisition unit 31 (S1).

[0044] The physical information estimation unit 32 extracts input data for a certain time interval from the data acquired by the data acquisition unit 31 (S2). In estimating the tire physical information, acceleration data for at least one axis (for example, in the circumferential direction) is required as input data. In estimating the tire physical information, acceleration data for two axes, for example, in the circumferential direction and the axial direction of the tire 10, may be used as input data, or acceleration data for three axes may be used as input data. Furthermore, time-series data for at least one of strain in the tire 10, tire pressure, and tire temperature may be included in the input data.

[0045] The feature extraction unit 51 of the computation model 32a executes a process of extracting features by performing convolution and pooling operations on the input data (S3). The full connection unit 53 of the computation model 32a executes a full connection operation on the features extracted by the feature extraction unit 51 and input to each node of the intermediate layer 52 (S4). Parameters such as weighting used in the full connection operation are determined during training of the computation model 32a. Normalized tire physical information such as tire force F, road friction coefficient, and moments acting on the tire 10 about three axes is output to each node of the output layer 54 by the full connection operation.

[0046] The physical information estimation unit 32 estimates the tire physical information by performing an inverse operation of the normalization on the normalized tire physical information output to the output layer 54 to restore the tire physical information (S5), and then ends the process.

[0047] 6(a) to 6(c) are graphs showing the correlation between the estimated value and the measured value by the tire physical information estimation system 100. Fig. 6(a) shows the correlation between the estimated value and the measured value of the tire force Fx, Fig. 6(b) shows the correlation between the estimated value and the measured value of the tire force Fy, and Fig. 6(c) shows the correlation between the estimated value and the measured value of the tire force Fz.

[0048] Figures 7(a) to 7(c) are graphs showing the correlation between estimated values ​​and measured values ​​when a learning model of a comparative example is used. The correlation between estimated values ​​and measured values ​​of tire forces Fx, Fy, and Fz in the comparative example is shown in Figures 7(a), 7(b), and 7(c), respectively. In the comparative example, a single-task learning model (referred to as the "conventional model") was constructed. The conventional model provides high estimation accuracy but requires a large amount of calculation, making it unsuitable for installation in a vehicle.

[0049] The correlation of the tire force Fx estimated by the tire physical information estimation system 100 according to the present embodiment shown in Fig. 6(a) has a distribution that is generally equivalent to the correlation of the tire force Fx according to the comparative example shown in Fig. 7(a). Similarly, when comparing Fig. 6(b) with Fig. 7(b) and Fig. 6(c) with Fig. 7(c), the correlation of the tire forces Fy and Fz estimated by the tire physical information estimation system 100 according to the present embodiment has a distribution that is generally equivalent to the correlation of the tire forces Fy and Fz according to the comparative example.

[0050] FIG. 8 is a chart showing the mean absolute errors of estimated and measured values. The mean absolute errors for tire forces Fx, Fy, and Fz are roughly equivalent between the multi-task model of this embodiment and the conventional model of the comparative example. This shows that the multi-task learning model with normalized output of this embodiment can achieve the same estimation accuracy of tire force F as the conventional model of the comparative example.

[0051] The calculation model 32a of the tire physical information estimation system 100 can improve the estimation accuracy of the tire physical information by being configured to output normalized tire physical information in at least two axial directions from the output layer 54. As described above, the calculation model 32a normalizes the vertical tire force Fz using a calculation different from that used for the other two axial tire forces Fx and Fy, thereby making it possible to improve the estimation accuracy by assuming that the normalized tire forces vary within the same range.

[0052] The tire physical information estimation system 100 can reduce the scale of calculations by providing a full connection unit 53 for each of the multiple tire physical information to be output in the calculation model 32a and multitasking the calculation model 32a. The feature extraction unit 51 of the calculation model 32a reduces the scale of calculations by sharing filters in the convolution calculations 51a and 51c. The physical information estimation unit 32 estimates, for example, at least two axial tire forces F out of three axial tire forces F as tire physical information using the calculation model 32a. Furthermore, by estimating all three axial tire forces F, the physical information estimation unit 32 can provide information necessary for analyzing behaviors of the tire 10, such as slippage.

[0053] The computation model generation system 110 normalizes the tire physical information measured by the tire physical information measurement device 60 and uses the normalized information as training data to train the computation model 32a, thereby generating a computation model 32a with good estimation accuracy. The learning processing unit 71 of the computation model generation system 110 trains the multitasking computation model 32a, thereby generating a computation model with a low computation scale.

[0054] (Variation) Fig. 9 is a block diagram showing a functional configuration of a tire physical information estimation system 100 according to a modified example. In the modified example shown in Fig. 9, data to be input to the calculation model 32a is acquired from a vehicle control device 90. Note that both the data from the vehicle control device 90 and the data from the sensor 20 (see Fig. 2) can also be used as data to be input to the calculation model 32a.

[0055] The vehicle control device 90 acquires driving data such as the vehicle's driving speed, acceleration in three axial directions, and three-axial angular velocity from a digital tachometer or the like of the vehicle, as well as load data such as the vehicle weight and axle loads on the axles. The vehicle control device 90 outputs the driving data and load data to the tire physical information estimation device 30.

[0056] The tire physical information estimation device 30 estimates tire physical information such as the tire force F, the road friction coefficient, and moments around three axes acting on the tire 10 using a calculation model 32a based on data input from a vehicle control device 90. The calculation model 32a is constructed by performing learning in advance to estimate tire physical information based on data input from the vehicle control device 90, for example, through test driving using an actual vehicle.

[0057] In the above-described embodiment and modified examples, tire physical information estimated by the computational model 32a has been described as including tire force F, road friction coefficient, and moments acting on the tire 10 about three axes. However, it is also possible to estimate, for example, loosening of fastening parts such as wheel nuts used to mount the tire 10. Since vibrations due to loosening of fastening parts such as wheel nuts appear in the acceleration data measured on the tire 10, a computational model 32a that estimates loosening of fastening parts by comparison with other tire forces F is constructed and trained in a CNN format. The tire physical information estimation system 100 executes calculations using the computational model 32a based on input data such as acceleration data acquired during actual vehicle travel, and can estimate loosening of fastening parts of the tire 10 in real time.

[0058] 1, the sensor 20 may be, for example, a microphone provided on the tire 10 or in the vicinity of the tire 10. The calculation model 32a may estimate the tire physical information using audio data collected by a microphone or the like.

[0059] In the above-described embodiment and modified examples, the computational model 32a uses a CNN-type LeNet model, but it is also possible to use a model structure such as a so-called DenseNet model, ResNet model, MobileNet model, or PeleelNet model. Furthermore, a model may be constructed by incorporating a modular structure such as a Dense Block, Residual Block, or Stem Block into the computational model 32a.

[0060] Next, features of the tire physical information estimation system 100 and the computational model generation system 110 according to the embodiment will be described. A tire physical information estimation system 100 according to the embodiment includes a physical information estimation unit 32 and a data acquisition unit 31. The physical information estimation unit 32 has a learning-type computational model 32a extending from an input layer 50 to an output layer 54, and estimates tire physical information resulting from the motion of the tire 10. The data acquisition unit 31 acquires input data for the input layer 50. The computational model 32a has a feature extraction unit 51 that executes convolution operations 51a and 51c during intermediate operations from the input layer 50 to the output layer 54, and outputs normalized tire physical information in at least two axial directions from the output layer 54. This enables the tire physical information estimation system 100 to improve the accuracy of estimation of tire physical information such as tire force F, road friction coefficient, and moments acting on the tire 10 about three axes.

[0061] The computational model 32a also has a plurality of full-connection units 53 corresponding to each axial direction, and inputs the output of the feature extraction unit 51 to the plurality of full-connection units 53. As a result, the tire physical information estimation system 100 can reduce the scale of computation by multitasking the computational model 32a.

[0062] The tire physical information is the tire forces F in the three axial directions. This allows the tire physical information estimation system 100 to provide information necessary for analyzing the behavior of the tire 10, such as slippage.

[0063] Furthermore, the output layer 54 of the computational model 32a outputs the vertical tire force Fz normalized by a different calculation from the other two axial tire forces Fx and Fy. This allows the tire physical information estimation system 100 to assume that the normalized tire forces vary within the same range, thereby improving estimation accuracy.

[0064] The computational model generation system 110 includes a physical information estimation unit 32, a data acquisition unit 31, and a learning processing unit 71. The physical information estimation unit 32 has a learning-type computational model 32a extending from an input layer 50 to an output layer 54, and estimates tire physical information generated by the motion of the tire 10. The data acquisition unit 31 acquires input data for the input layer 50. The learning processing unit 71 trains the computational model 32a based on training data obtained by normalizing tire physical information measured on the tire 10. The computational model 32a has a feature extraction unit 51 that executes convolution operations 51a and 51c during intermediate operations from the input layer 50 to the output layer 54, and outputs normalized tire physical information in at least two axial directions from the output layer 54. This enables the computational model generation system 110 to generate a computational model 32a with high estimation accuracy.

[0065] The present invention has been described above based on the embodiments. These embodiments are merely examples, 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 also fall within the scope of the claims of the present invention. Therefore, the descriptions and drawings in this specification should be treated as illustrative rather than restrictive. [Explanation of symbols]

[0066] 10 tire, 31 data acquisition unit, 32 physical information estimation unit, 32a computational model, 50 input layer, 51 feature extraction unit, 51a, 51c convolution operation, 53 fully connected part, 54 output layer, 71 learning processing unit, 100 tire physical information estimation system, 110 Computational Model Generation System.

Claims

1. a physical information estimation unit having a learning-type calculation model extending from an input layer to an output layer, and estimating tire physical information generated by tire motion; a data acquisition unit that acquires input data for the input layer, the computational model includes a feature extraction unit that executes a convolution operation in an intermediate operation from the input layer to the output layer, and outputs normalized tire forces in three axial directions as the tire physical information from the output layer; The tire physical information estimation system is characterized in that the output layer outputs a vertical tire force normalized by a calculation different from that for the tire forces in the other two axial directions.

2. The tire physical information estimation system according to claim 1, wherein the calculation model has a plurality of full-connection sections corresponding to each axial direction, and the output of the feature extraction section is input to the plurality of full-connection sections.

3. a physical information estimation unit having a learning-type calculation model extending from an input layer to an output layer, and estimating tire physical information generated by tire motion; a data acquisition unit that acquires input data for the input layer; a learning processing unit that learns the calculation model based on teacher data obtained by normalizing the tire physical information measured on the tire, the computational model includes a feature extraction unit that executes a convolution operation in an intermediate operation from the input layer to the output layer, and outputs normalized tire forces in three axial directions as the tire physical information from the output layer; The calculation model generation system is characterized in that the output layer outputs a vertical tire force normalized by a calculation different from that for the tire forces in the other two axial directions.

Images