JP7493647B2 - Wear amount estimation device and wear amount estimation method - Google Patents

Description

The present invention relates to a tire wear amount calculation model generation system, a wear amount estimation system, and a calculation model generation method.

In general, tires wear down depending on the driving conditions and distance traveled. Recently, devices have been commercialized that attach sensors to tires to measure tire pressure and temperature and display the measured pressure and temperature.

Patent document 1 describes a conventional tire wear estimation system. The tire wear estimation system has at least one sensor attached to a tire to generate a first predictive variable, at least one of a lookup table and a database that stores data related to a second predictive variable, one of the predictive variables including at least one vehicle effect, and a model that receives the predictive variables and generates an estimated wear rate for at least one tire.

JP 2018-158722 A

The tire wear estimation system described in Patent Document 1 estimates tire wear using, for example, wheel position and drive train as vehicle influences. Tire wear due to wheel position and drive train has a greater or lesser influence depending on the type of vehicle, and is not necessarily a major factor in large vehicles such as trucks, for example. Therefore, the inventors realized that there is room for improvement in tire wear estimation. In other words, when estimating the amount of tire wear, it is necessary to take into account that factors such as uneven loading weight on a vehicle can significantly change the load on each tire, affecting the amount of wear.

The present invention was made in consideration of the above circumstances, and its purpose is to provide a calculation model generation system, a wear amount estimation system, and a calculation model generation method that can accurately estimate tire wear.

One aspect of the present invention is a calculation model generation system. It includes a tire load calculation unit that calculates the tire load applied to multiple tires mounted on a vehicle, a wear amount calculation unit that has a calculation model that calculates the tire wear amount based on the load and inputs the tire load calculated by the tire load calculation unit to calculate the tire wear amount using the calculation model, and a calculation model update unit that compares the wear amount measured on the tire with the wear amount calculated by the wear amount calculation unit and updates the calculation model.

Another aspect of the present invention is a wear amount estimation system. It includes a tire load calculation unit that calculates the tire load applied to multiple tires mounted on a vehicle, and a wear amount calculation unit that has a calculation model that calculates the tire wear amount based on the load, and inputs the tire load calculated by the tire load calculation unit and calculates the wear amount of the tire using the calculation model.

Another aspect of the present invention is a calculation model generation method. The calculation model generation method includes a tire load calculation step of calculating tire loads applied to multiple tires mounted on a vehicle, a wear amount calculation step of inputting the tire load calculated in the tire load calculation step based on a calculation model that calculates tire wear amount based on the load and calculating the wear amount of the tire using the calculation model, and a calculation model update step of comparing the wear amount measured on the tire with the wear amount calculated in the wear amount calculation step and updating the calculation model.

The present invention allows for accurate estimation of tire wear.

FIG. 1 is a schematic diagram for explaining an overview of a tire wear amount calculation model generation system. 1 is a block diagram showing a functional configuration of a computation model generation system according to a first embodiment. FIG. FIG. 11 is a schematic diagram for explaining learning of a computation model. 1 is a flowchart showing a procedure for generating a computation model by the computation model generation system. FIG. 11 is a block diagram showing a functional configuration of a wear amount estimation system according to a second embodiment. FIG. 13 is a schematic diagram for explaining a modified example for measuring unevenness of a load on a truck or the like.

The present invention will be described below based on a preferred embodiment with reference to Figures 1 to 6. The same or equivalent components and parts shown in each drawing are given the same reference numerals, and duplicated explanations will be omitted as appropriate. Furthermore, the dimensions of the parts in each drawing are enlarged or reduced as appropriate to facilitate understanding. Furthermore, some of the parts that are not important for explaining the embodiment are omitted in each drawing.

(Embodiment 1)
1 is a schematic diagram for explaining an overview of a tire wear amount computation model generation system 100. A vehicle 8 is equipped with a plurality of tires 10, and the computation model generation system 100 generates a computation model 34a that calculates the amount of tire wear for each tire 10, based on various information measured on the vehicle 8. The computation model generation system 100 calculates the tire load acting on each tire 10, based on load data measured by a plurality of load sensors arranged on the vehicle 8.

The computational model generation system 100 may also determine the tire load applied to each tire 10 based on vehicle-side data such as axle load and acceleration provided by the vehicle 8, and tire-side data from acceleration sensors and pressure sensors attached to the tires. Furthermore, the computational model generation system 100 may determine the tire load applied to each tire 10 by taking into account the inclination of the vehicle 8 obtained from the position data of the vehicle 8.

The computational model generation system 100 calculates the amount of tire wear using the computational model 34a based on the tire load. The computational model generation system 100 uses a learning model such as a neural network as the computational model, and uses the amount of tire wear actually measured on the tire 10 as training data, and repeats learning by performing calculations and updating the computational model to improve the accuracy of the computational model 34a. The wear amount estimation device 30 functions as a device that estimates the amount of tire wear after the computational model has learned. For example, when there are four grooves in the width direction, the tire wear amount is measured by measuring the groove depth of the number of grooves and three points at 120° intervals in the circumferential direction of each groove. As a result, data on uneven wear in the circumferential direction as well as uneven wear in the width direction is input as training data, and a computational model that estimates uneven wear can be generated.

The computational model generation system 100 can perform computational model learning for a tire 10 of a certain specification by running a vehicle equipped with the tire 10 (including the wheel). The tire specifications include information on tire performance, such as the manufacturer, product name, tire size, tire width, aspect ratio, tire strength, static stiffness, dynamic stiffness, tire outer diameter, load index, and date of manufacture.

The computational model generation system 100 may construct a computational model 34a that uses as input the air pressure, temperature, and measurement data of the acceleration occurring in the tire 10. The acceleration occurring in the tire 10 is, for example, triaxial acceleration in the radial, axial, and front-to-rear directions of the tire 10, and this triaxial acceleration is input to the computational model 34a. Also, for example, biaxial acceleration in the radial and axial directions of the tire 10 may be input to the computational model 34a.

Figure 2 is a block diagram showing the functional configuration of the computational model generation system 100 according to the first embodiment. The computational model generation system 100 includes a load sensor 21, a wear amount estimation device 30, and a tire measurement device 40. The load sensor 21 is, for example, a load cell, and is arranged at multiple locations on the vehicle. If the vehicle 8 is a truck, the load sensor 21 may be arranged between the vehicle body and the loading platform. In addition, the load sensor 21 is arranged, for example, on a suspension provided between the axle and the vehicle body in order to obtain the tire load applied to the tire 10. Many vehicles, not limited to trucks, employ suspensions to reduce vibrations from the road surface, making the system highly versatile. The load sensor 21 may be a device that calculates the load based on the dynamic characteristics of the suspension.

The tire measuring device 40 measures the depth of the grooves in the tread of the tire 10. An operator may measure the groove depth using a measuring tool, a camera, or visually, and the tire measuring device 40 may store the measurement data entered by the operator. The tire measuring device 40 may also be a dedicated device that measures and stores the groove depth using a mechanical or optical method.

The tire 10 is provided with tire-mounted sensors 20, such as a pressure gauge, a temperature sensor, and an acceleration sensor. The pressure gauge and temperature sensor are provided, for example, in the air valve of the tire 10, and measure the air pressure and temperature of the tire 10, respectively. The temperature sensor may be provided in the inner liner of the tire 10, for example, to accurately measure the temperature of the tire 10. The acceleration sensor is provided, for example, in the tire 10 and wheel, for example, and measures the acceleration occurring in the tire 10. Note that the tire 10 may be provided with an RFID or the like to which unique identification information is assigned, in order to identify each tire.

The wear amount estimation device 30 has a tire load calculation unit 31, a vehicle information acquisition unit 32, a position information acquisition unit 33, a wear amount calculation unit 34, and a calculation model update unit 35. The wear amount estimation device 30 is an information processing device such as a PC (personal computer). Each unit in the wear amount estimation device 30 can be realized in hardware terms by electronic elements and mechanical parts such as a computer CPU, and in software terms by a computer program, but here, functional blocks realized by the cooperation of these are depicted. 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.

The tire load calculation unit 31 calculates the tire load based on the measurement data from the load sensor 21. For example, the tire load calculation unit 31 determines the center position of the load distribution based on the measurement data from the load sensor 21, and determines the tire load occurring at the position of each tire 10 mounted on the vehicle 8.

The tire load calculation unit 31 may calculate the tire load based on vehicle data including the axle load and three-axle acceleration acquired from the vehicle 8, without relying on the measurement data from the load sensor 21. The vehicle information acquisition unit 32 acquires vehicle data including the axle load and three-axle acceleration of the vehicle 8 from the digital tachograph 23 mounted on the vehicle 8, and outputs the data to the tire load calculation unit 31. The tire load calculation unit 31 calculates the movement of the center of gravity of the vehicle 8 based on the axle load and three-axle acceleration, and calculates the tire load generated on each tire 10.

The tire load calculation unit 31 may also calculate the inclination of the vehicle 8 calculated from the position data of the vehicle 8, and calculate the tire load generated on each tire 10 by taking into account vehicle side data such as the axle load and three-axle acceleration. The position information acquisition unit 33 acquires position data from the GPS receiver 24 mounted on the vehicle 8, and outputs it to the tire load calculation unit 31. The tire load calculation unit 31 can determine the inclination of the vehicle 8, for example, from the change over time in altitude information included in the position data.

The wear amount calculation unit 34 has a calculation model 34a, and input data such as the tire load calculated by the tire load calculation unit 31 is input to the calculation model 34a to calculate the tire wear amount. FIG. 3 is a schematic diagram for explaining the learning of the calculation model 34a. The input data to the calculation model 34a includes tire load data calculated by the tire load calculation unit 31 and tire data including the air pressure, temperature, and acceleration of the tire 10. In addition to these, the input data may include weather data and information on road surface conditions. Weather data such as the temperature and precipitation in the driving area are used as input data to the calculation model 34a. Information on road surface conditions such as unevenness, temperature, and dryness/wetness of the road surface on which the vehicle is traveling are used as input data to the calculation model 34a.

The computation model 34a uses a learning model such as a neural network. The computation model 34a inputs tire load and tire data (tire air pressure, temperature, acceleration, etc.) to the nodes of the input layer, and performs computations using paths from the input layer with weighting to the intermediate layer. The computation model 34a further performs computations using paths with weighting from the intermediate layer to the output layer, and outputs the amount of tire wear from the nodes of the output layer. In a learning model such as a neural network, in addition to linear computations, nonlinear computations may be performed using activation functions, etc.

The calculation model update unit 35 compares the tire wear amount as the calculation result with the teacher data to update the calculation model 34a. The calculation model update unit 35 has a wear amount comparison unit 35a and an update processing unit 35b. The wear amount comparison unit 35a compares the tire wear amount calculated by the calculation model 34a with the tire wear amount measured by the tire measuring device 40 as the teacher data, and outputs the error to the update processing unit 35b.

The update processing unit 35b updates the weighting of the paths on the calculation model based on the wear amount error calculated by the calculation model 34a. The accuracy of the calculation model is improved by repeating the calculation of the tire wear amount by the calculation model 34a, the comparison with the teacher data by the wear amount comparison unit 35a, and the update of the calculation model by the update processing unit 35b. In addition, the update processing unit 35b may determine that the learning of the calculation model 34a is completed when the estimated value of the tire wear amount and the measured tire wear amount, which is the teacher data, are within a predetermined error range, and end the update of the calculation model 34a. For example, it may be determined that the learning is completed when the error in the groove depth of the tire 10 as the tire wear amount is within 0.1 mm, or when the estimated value of the tire wear amount is within 10% above or below the measured tire wear amount, but the error range as a judgment criterion is not limited to these numerical values.

Next, the operation of the computational model generation system 100 will be described. FIG. 4 is a flowchart showing the procedure for generating a computational model by the computational model generation system 100. The wear amount estimation device 30 starts acquiring load data and the like from the load sensor 21 in the tire load calculation unit 31 (S1). In step S1, the tire load calculation unit 31 may acquire vehicle-side data such as axle load and three-axle acceleration from the vehicle information acquisition unit 32 instead of the load data from the load sensor 21, and may further acquire position data from the position information acquisition unit 33.

The tire load calculation unit 31 calculates the tire load applied to each tire 10 based on the acquired data (S2). The tire load calculation unit 31 calculates and accumulates the tire load over a predetermined period (S3). The predetermined period may be, for example, a period during which the vehicle travels once, or may be several days or a week, but is not limited thereto. After the predetermined period has elapsed, the wear amount calculation unit 34 inputs the accumulated tire load to the calculation model 34a and estimates the tire wear amount (S4). The wear amount comparison unit 35a compares the tire wear amount calculated by the calculation model 34a with the tire wear amount measured by the tire measuring device 40 as teacher data (S5). The wear amount comparison unit 35a outputs the error between the tire wear amount calculated by the calculation model 34a and the tire wear amount measured by the tire measuring device 40 as a comparison result to the update processing unit 35b.

The update processing unit 35b updates the calculation model based on the error in the tire wear amount input from the wear amount comparison unit 35a (S6), and ends the process. By repeating these processes, the wear amount estimation device 30 updates the calculation model and improves the accuracy of the tire wear amount estimation.

The computational model generation system 100 can generate a computational model 34a for multiple tires 10 mounted on a vehicle 8, using the tire load applied to each tire 10 as input data. The computational model generation system 100 can generate a computational model that accurately estimates the amount of tire wear by taking into account the tire load on each tire 10, which changes depending on the position and weight bias of the load, the physical state of the load (solid, liquid, powder), and the running state of the vehicle, i.e., the acceleration and inclination in three axial directions.

In addition, by arranging load sensors 21 at multiple locations on the vehicle 8, the computational model generation system 100 can accurately calculate tire loads based on load data measured by the multiple load sensors 21. In addition, by arranging load sensors 21 on the suspension, the computational model generation system 100 can be applied to many vehicles equipped with suspensions, making it highly versatile.

The computational model generation system 100 can calculate the tire load on each tire 10 according to the running state of the vehicle 8 by acquiring vehicle side data such as axle load and three-axis acceleration from the vehicle information acquisition unit 32 instead of load data from the load sensor 22. In addition, the computational model generation system 100 can further acquire position data from the position information acquisition unit 33 and calculate the tire load more accurately by taking into account the inclination of the vehicle 8 calculated from the position data.

The computation model generation system 100 can generate a computation model that estimates the amount of tire wear more accurately by inputting the tire data (air pressure, temperature, acceleration, etc. of the tire 10), meteorological data, road surface conditions, etc. shown in FIG. 2 into the computation model 34a in addition to the tire load calculated by the tire load calculation unit 31. Note that the computation model generation system 100 may accumulate input data to the computation model 34a, such as meteorological data for the vehicle's traveling position and road surface conditions, for a predetermined period of time, similar to the tire load.

(Embodiment 2)
5 is a block diagram showing a functional configuration of a wear amount estimation system 110 according to embodiment 2. After the computation model 34a for estimating a tire wear amount is generated by the computation model generation system 100 according to embodiment 1 as described above, a wear amount estimation device 30 including the computation model 34a can be configured. The wear amount estimation system 110 includes a load sensor 21 and the like and the wear amount estimation device 30, and uses the computation model 34a generated by the computation model generation system 100 to accurately estimate the wear amount of a tire 10 mounted on a vehicle.

The wear amount estimation device 30 includes a tire load calculation unit 31, a vehicle information acquisition unit 32, a position information acquisition unit 33, a wear amount calculation unit 34, and a notification unit 36. The wear amount estimation device 30 may be mounted on the vehicle 8, or may not be mounted on the vehicle 8 and may estimate the tire wear amount by acquiring various data from the vehicle 8 via a communication connection. The tire load calculation unit 31 , the vehicle information acquisition unit 32, and the position information acquisition unit 33 have the same configurations and operations as those in the first embodiment, and therefore descriptions thereof will be omitted for brevity.

The wear amount calculation unit 34 uses the calculation model generated by the calculation model generation system 100 according to the first embodiment as the calculation model 34a. The wear amount calculation unit 34 may calculate the tire wear amount after the tire load calculation unit 31 calculates the tire load and accumulates it for a predetermined period of time, or may calculate the tire wear amount from time to time at the acquired timing.

The notification unit 36 notifies the amount of tire wear by displaying on the display device 51 or outputting audio from the speaker 52 so that passengers such as the driver of the vehicle and operation managers who manage the vehicles at transport companies and the like can be informed of the current amount of tire wear. The notification unit 36 may also notify the current amount of tire wear to a vehicle control device 53 mounted on the vehicle. The vehicle control device 53 can perform vehicle control such as automatic driving and collision avoidance based on the current amount of tire wear.

(Modification)
6 is a schematic diagram for explaining a modified example for measuring uneven distribution of cargo on a truck or the like. In this modified example, a monitoring device 25 for monitoring cargo, such as a camera or radar, is provided on the bed of a vehicle 8. The cargo placement is monitored by the monitoring device 25, and a wear amount estimation device 30 determines the distribution of the cargo carried and weights the tire load for each tire 10 in relation to the tire position. The wear amount estimation device 30 can determine the tire load according to the weighting for each tire 10.

Next, features of the computational model generation system 100, the wear amount estimation system 110, and the computational model generation method according to each embodiment and modification will be described.
The computation model generation system 100 includes a tire load calculation unit 31, a wear amount calculation unit 34, and a computation model update unit 35. The tire load calculation unit 31 calculates the tire load applied to a plurality of tires 10 mounted on a vehicle 8. The wear amount calculation unit 34 has a computation model 34a that calculates the tire wear amount based on the load, and inputs the tire load calculated by the tire load calculation unit 31 to calculate the wear amount of the tire 10 using the computation model 34a. The computation model update unit 35 compares the wear amount measured on the tire 10 with the wear amount calculated by the wear amount calculation unit 34, and updates the computation model 34a. This allows the computation model generation system 100 to generate a computation model that accurately estimates the tire wear amount.

The computational model generation system 100 also includes load sensors 21 provided at multiple locations on the vehicle 8. The tire load calculation unit 31 calculates the tire load based on the load data measured by the load sensors 21. This allows the computational model generation system 100 to accurately calculate the tire load based on the load data measured by the multiple load sensors 21.

The load sensor 21 is also provided on the suspension of the vehicle 8. This allows the computational model generation system 100 to be applied to many vehicles equipped with suspensions, making it highly versatile.

The computational model generation system 100 also includes a vehicle information acquisition unit 32 that acquires vehicle-side data including axle load and acceleration from the vehicle 8. The tire load calculation unit 31 calculates the tire load based on the vehicle-side data acquired by the vehicle information acquisition unit 32. This allows the computational model generation system 100 to calculate the tire load on each tire 10 according to the running state of the vehicle 8.

The computational model generation system 100 also includes a position information acquisition unit 33 that acquires vehicle position data. The tire load calculation unit 31 calculates the tire load based on the vehicle side data acquired by the vehicle information acquisition unit 32 and the inclination of the vehicle 8 calculated from the position data acquired by the position information acquisition unit 33. This allows the computational model generation system 100 to more accurately calculate the tire load by taking into account the inclination of the vehicle 8 calculated from the position data.

In addition, at least one of weather data and road surface conditions is input to the wear amount calculation unit 34. This allows the computational model generation system 100 to generate a computational model that takes into account the weather data or road surface conditions to more accurately estimate the amount of tire wear.

The wear amount estimation system 110 includes a tire load calculation unit 31 and a wear amount calculation unit 34. The tire load calculation unit 31 calculates the tire load applied to a plurality of tires 10 mounted on the vehicle 8. The wear amount calculation unit 34 has a calculation model 34a that calculates the tire wear amount based on the load, and inputs the tire load calculated by the tire load calculation unit 31 to calculate the wear amount of the tire 10 using the calculation model 34a. In this way, the wear amount estimation system 110 accurately estimates the wear amount of the tire 10 mounted on the vehicle using the calculation model 34a generated by the calculation model generation system 100.

The computation model generation method includes a tire load calculation step, a wear amount calculation step, and a computation model update step. The tire load calculation step calculates the tire load applied to a plurality of tires 10 mounted on a vehicle 8. The wear amount calculation step inputs the tire load calculated in the tire load calculation step based on a computation model 34a that calculates the tire wear amount based on the load, and calculates the wear amount of the tire 10 using the computation model 34a. The computation model update step compares the wear amount measured on the tire 10 with the wear amount calculated in the wear amount calculation step, and updates the computation model 34a. According to this computation model generation method, a computation model that accurately estimates the tire wear amount can be generated.

The above describes the embodiments of the present invention. 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 are also 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.

10 Tire, 31 Tire load calculation unit, 32 Vehicle information acquisition unit,
33 position information acquisition unit, 34 wear amount calculation unit, 34a calculation model,
35 Calculation model update unit, 8 Vehicle,
100 computational model generation system, 110 wear amount estimation system.

Claims (6)

A vehicle information acquisition unit that acquires vehicle side data from a vehicle;
a tire load calculation unit that calculates tire loads applied to a plurality of tires mounted on a vehicle;
a wear amount calculation unit having a learning type calculation model that outputs a wear amount for each groove of the tire based on input data, and inputting tire data including the vehicle side data, the tire load, and the temperature and pressure of the tire into the calculation model to calculate the wear amount for each groove of the tire using the calculation model;
The wear amount calculation unit calculates the wear amount of each groove of the tire using the trained computation model ,
The tire load calculation unit weights the tire load for each tire in relation to the tire position by a monitoring device for a loading platform of the vehicle,
The wear amount calculation unit is a wear amount estimation device that calculates the wear amount for each groove of the tire based on the weighting of the tire load . A vehicle information acquisition unit that acquires vehicle side data from a vehicle;
a tire load calculation unit that calculates tire loads applied to a plurality of tires mounted on a vehicle;
a wear amount calculation unit having a learning type calculation model that outputs a wear amount for each groove of the tire based on input data, and inputting tire data including the vehicle side data, the tire load, and the temperature and pressure of the tire into the calculation model to calculate the wear amount for each groove of the tire using the calculation model;
The wear amount calculation unit calculates the wear amount of each groove of the tire using the trained computation model ,
A location information acquisition unit that acquires location data of the vehicle is further provided.
The tire load calculation unit calculates the tire load based on the longitudinal inclination of the vehicle calculated based on the position data acquired by the position information acquisition unit . 3. The wear amount estimating device according to claim 1, further comprising an informing unit that informs of a current amount of wear of a tire. 3. The wear amount estimating device according to claim 1 , wherein at least one of weather data and road surface conditions is input to the wear amount calculation unit. A vehicle information acquisition step of acquiring vehicle side data from a vehicle;
a tire load calculation step of calculating tire loads applied to a plurality of tires mounted on a vehicle;
a wear amount calculation step of calculating the wear amount for each groove of the tire using tire data including the vehicle side data, the tire load, and the temperature and pressure of the tire, based on a learning type calculation model that outputs the wear amount for each groove of the tire based on input data,
the wear amount calculation step calculates a wear amount for each groove of the tire using the trained computation model ,
The tire load calculation step weights the tire load of each tire in relation to the tire position by a monitoring device of a loading platform of the vehicle,
The wear amount calculation step calculates the amount of wear for each groove of the tire based on a weighting of the tire load . A vehicle information acquisition step of acquiring vehicle side data from a vehicle;
a tire load calculation step of calculating tire loads applied to a plurality of tires mounted on a vehicle;
a wear amount calculation step of calculating the wear amount for each groove of the tire using tire data including the vehicle side data, the tire load, and the temperature and pressure of the tire, based on a learning type calculation model that outputs the wear amount for each groove of the tire based on input data,
the wear amount calculation step calculates a wear amount for each groove of the tire using the trained computation model,
A position information acquisition step of acquiring position data of the vehicle is further provided,
The tire load calculation step calculates the tire load based on a longitudinal tilt of the vehicle calculated based on the position data acquired in the position information acquisition step.

Images