CN116262403A

CN116262403A - Tire stiffness estimation system

Info

Publication number: CN116262403A
Application number: CN202211612194.1A
Authority: CN
Inventors: A·诺里; K·B·辛格
Original assignee: Goodyear Tire and Rubber Co
Current assignee: Goodyear Tire and Rubber Co
Priority date: 2021-12-15
Filing date: 2022-12-15
Publication date: 2023-06-16

Abstract

A tire longitudinal stiffness estimation system includes an electronic communication system disposed on a vehicle. A sensor is disposed on the vehicle in communication with the electronic communication system and a processor is accessible through the electronic communication system. A sensor measures a parameter associated with the vehicle and communicates data of the parameter to the processor. A mu slip curve generator receives the parameters to generate a mu slip curve in real-time from the data. An extraction module extracts raw data from the linear portion of the mu slip curve. A denoising module denoises the raw data from the mu slip curve by determining a vector of the raw data, an orientation of the vector, and an orientation of the vector. The denoising module generates denoising data, and the stiffness calculator receives the denoising data and generates a longitudinal stiffness estimation for the tire.

Description

Tire stiffness estimation system

Technical Field

The present invention relates generally to tire monitoring and estimation systems. More particularly, the present invention relates to systems for predicting certain tire characteristics. In particular, the present invention relates to a system for estimating the stiffness of a tire in real time.

Background

As is known in the art, a vehicle is supported by a plurality of tires. The stiffness of each tire during operation of the vehicle can affect the performance and characteristics of the tire. For example, the longitudinal stiffness of the tire (which is the stiffness of the tire in its longitudinal or running direction) may be employed to distinguish between different road conditions and/or different wear states of the tire. Further, longitudinal stiffness may be employed to improve operation of vehicle control systems, such as Adaptive Cruise Control (ACC), antilock Braking System (ABS), electronic Stability Program (ESP), acceleration Slip Regulation (ASR), and the like.

Because of the usefulness of the tire longitudinal stiffness, it is desirable to generate an accurate estimate of the longitudinal stiffness. In the prior art, systems have been developed that provide such estimates. However, such prior art systems have been complex in order to achieve accurate longitudinal stiffness estimates, often employing data from multiple sources. For example, data from vehicles, from tires, and from a remote data server may be used.

Using such complex systems and data from such various sources may be undesirably difficult to achieve. Furthermore, such complex systems require a considerable computational load. When a significant amount of computation load is involved, such systems may not be able to be executed on the vehicle-mounted processor, thereby undesirably requiring additional sources, such as cloud computing, and undesirably spending a significant amount of time generating real-time estimates.

Accordingly, there is a need in the art for a system for estimating tire longitudinal stiffness in real time that provides accurate estimates based on data from limited sources and has a low computational load.

Disclosure of Invention

According to an aspect of an exemplary embodiment of the present invention, a longitudinal stiffness estimation system for at least one tire of a supporting vehicle is provided. The system comprises: an electronic communication system disposed on the vehicle; and at least one sensor disposed on the vehicle in electronic communication with the electronic communication system. The processor is accessible through the electronic communication system. A sensor measures a selected parameter associated with the vehicle and communicates data of the selected parameter to the processor through the electronic communication system. A mu slip curve generator communicates with the processor, receives the selected parameters, and generates a mu slip curve in real-time from the transmitted data. An extraction module communicates with the processor and extracts raw data from the linear portion of the mu slip curve. A denoising module communicates with the processor and denoises the raw data from the mu slip curve by determining a vector of the raw data, an orientation of the vector, and an orientation of the vector. The denoising module generates denoising data, and the stiffness calculator receives the denoising data and generates a longitudinal stiffness estimation for the tire.

Drawings

The invention will be described by way of example and with respect to the accompanying drawings in which:

FIG. 1 is a perspective view of a tire and vehicle employed in connection with the longitudinal stiffness estimation system of the present invention;

FIG. 2 is a flow chart illustrating an exemplary embodiment of a longitudinal stiffness estimation system of the present invention;

FIG. 3 is a graphical representation of a portion of the longitudinal stiffness estimation system shown in FIG. 2;

FIG. 4 is a graphical representation of another portion of the longitudinal stiffness estimation system shown in FIG. 2;

FIG. 5 is a graphical representation of another portion of the longitudinal stiffness estimation system shown in FIG. 2;

FIG. 6 is a graphical representation of another portion of the longitudinal stiffness estimation system shown in FIG. 2;

FIG. 7 is a graphical representation of another portion of the longitudinal stiffness estimation system shown in FIG. 2;

FIG. 8 is a graphical representation of a pavement condition monitor employing the longitudinal stiffness estimation system shown in FIG. 2; and

FIG. 9 is a graphical representation of a tire wear monitor employing the longitudinal stiffness estimation system shown in FIG. 2.

Like numbers refer to like parts throughout the drawings.

Definition of the definition

An "ANN" or "artificial neural network" is an adaptive tool for nonlinear statistical data modeling that changes its structure during a learning phase based on external or internal information flowing through the network. An ANN neural network is a nonlinear statistical data modeling tool that is used to model complex relationships between inputs and outputs, or to discover patterns in data.

"axial" and "axially" mean lines or directions parallel to the axis of rotation of the tire.

"CAN" is an abbreviation for controller area network and is used in conjunction with a CAN bus, which is an electronic communication system on a vehicle.

"circumferential" means a line or direction extending along the perimeter of the surface of the annular tread of the tire perpendicular to the axial direction.

"cloud computing" means computer processing involving computing power and/or data storage distributed across multiple data centers, typically facilitated through access and communication using the internet.

"inboard" means the side of the tire closest to the vehicle when the tire is mounted on the wheel and the wheel is mounted on the vehicle.

A "kalman filter" is a set of mathematical equations that implement a predictor-corrector type estimator that is optimal in the sense that: that is, when certain assumed conditions are met, it minimizes the estimated error covariance.

"transverse" means an axial direction.

The "lunberger (Luenberger) observer" is a state observer or estimation model. A "state observer" is a system that provides an estimate of the internal state of a given actual system based on measurements of the actual system's inputs and outputs. It is typically computer implemented and provides the basis for many practical applications.

The "MSE" is an abbreviation for mean square error, i.e. the error between the measured and estimated signals minimized by the kalman filter.

"outboard" means the side of the tire furthest from the vehicle when the tire is mounted on the wheel and the wheel is mounted on the vehicle.

"radial" and "radially" mean directions radially toward the axis of rotation of the tire or away from the axis of rotation of the tire.

"TPMS" means a tire pressure monitoring system.

Detailed Description

An exemplary embodiment of the longitudinal stiffness estimation system of the present invention is indicated at 10 in fig. 1-9. As shown in fig. 1, the system estimates the longitudinal stiffness of the tire 12 supporting the vehicle 14. Although the vehicle 14 is depicted as a passenger car, the invention is not so limited. The principles of the present invention find application in other vehicle categories, such as commercial trucks, where the vehicle may be supported by more or fewer tires.

Each tire 12 is of conventional construction and is mounted on a wheel 16. Each tire 12 includes a pair of sidewalls 18 extending to a circumferential tread 20. Each tire 12 may be equipped with a sensor or transducer 24, which sensor or transducer 24 may be a Tire Pressure Monitoring (TPMS) module or sensor, and detects tire parameters such as the pressure within the tire cavity 20 and the tire temperature. The sensor 24 is preferably secured to the innerliner 22 of the tire 12 by suitable means, such as adhesive.

The tire 12 includes a longitudinal stiffness, which is its stiffness in the longitudinal or running direction. Turning to fig. 2, the longitudinal stiffness estimation system 10 calculates the longitudinal stiffness of the tire 12 by providing a longitudinal stiffness estimate 52. Aspects of the longitudinal stiffness estimation system 10 are preferably executed on a processor 26, the processor 26 being accessible through an on-board electronic communication system (such as a CAN bus system 28) that enables central communication between a plurality of vehicle sensors. The processor 26 may be a local processor installed on the vehicle 14 or may be a remote processor, such as a cloud computing processor.

The longitudinal stiffness estimation system 10 preferably provides a longitudinal stiffness estimate 52 for each tire 12 mounted on the drive wheel 16 of the vehicle 14. For example, in the front wheel drive vehicle 14, the system 10 generates a stiffness estimate 52 for each of the front tires 12. For convenience, the system 10 is described with respect to one tire 12, with the understanding that the estimate 52 is preferably provided for each tire 12 mounted on the drive wheel 16 of the vehicle 14.

The longitudinal stiffness estimation system 10 receives as input certain parameters measured by sensors mounted on the vehicle 14 and in electronic communication with the vehicle CAN bus system 28. Specifically, the CAN bus 28 transmits the longitudinal acceleration (a _x Or a _x ) 30, wheel speed 32, throttle or accelerator pedal position 34, brake pedal position 36, and vehicle reference speed 38 are electronically transmitted to an acceleration module 40 and mu slip profile generator 42. The vehicle reference speed 38 may be obtained from a Global Positioning System (GPS) or other reliable source of vehicle reference speed.

In the acceleration module 40, the accelerator pedal position 34 and the brake pedal position 36 are employed to confirm that the vehicle 14 is accelerating. For example, if the accelerator pedal position 34 is below a predetermined throttle threshold, or if the brake pedal position 36 is above a predetermined brake threshold, the system 10 determines that the vehicle 14 is not accelerating. When the vehicle 14 is not accelerating, the system 10 does not advance to the [ mu ] slip curve generator 42. If the accelerator pedal position 34 is greater than a predetermined throttle threshold, and/or the brake pedal position 36 is below a predetermined brake threshold, the system 10 determines that the vehicle 14 is accelerating and proceeds to a [ mu ] slip curve generator 42.

With further regard to fig. 3, the [ mu ] slip curve generator 42 generates a [ mu ] slip curve 44 in real time. On the vertical axis, the [ mu ] slip curve 44 plots the friction between the tire 12 and the surface on which the tire is travelling, represented by the friction coefficient [ mu ] (mu ]. The [ mu ] slip curve generator 42 approximates the coefficient of friction [ mu ] during vehicle acceleration using the longitudinal acceleration 30 of the vehicle 14. On the horizontal axis, the [ mu ] slip curve 44 plots a tire slip 46, which tire slip 46 is the relative motion between the tire 12 and the surface over which the tire is traveling. The [ slip ] curve generator 42 uses the wheel speed 32 and the vehicle reference speed 38 to calculate slip 46:

slip 46, percent (%) = percent (%)

In this way, the [ mu ] slip curve generator 42 of the longitudinal stiffness estimation system 10 uses the input signal from the vehicle CAN bus system 28 to generate the [ mu ] slip curve 44 in real-time.

The slope 48 of the linear portion 50 of the [ mu ] slip curve 44 corresponds to the longitudinal stiffness of the tire 12. As will be described in greater detail below, the longitudinal stiffness estimation system 10 extracts the longitudinal stiffness of the tire 12 and provides the longitudinal stiffness estimate 52 in an accurate manner.

Referring to fig. 2 and 4, the extraction module 54 extracts raw data 56 from the linear portion 50 of the [ mu ] slip curve 44. The raw data 56 includes signal noise, i.e., unwanted modifications in the data that occur during acquisition, storage, transmission, and/or processing. Thus, the raw data 56 must be denoised or cleaned to improve its accuracy. It is to be understood that the data shown in fig. 4-9 may include hypothetical representations as examples for the purpose of illustrating the principles of the present invention.

Turning now to fig. 2 and 5, the denoising module 58 denoises or cleans up the raw data 56. The denoising module 58 preferably applies Principal Component Analysis (PCA) to determine patterns from the raw data 56 for predictive analysis. For example, PCA identifies principal components or feature vectors that are characteristic vectors of the linear transformation of the raw data 56. In this manner, the pattern visualized or represented by vector 60 is determined from raw data 56.

The PCA of the denoising module 58 predicts an orientation 62 of the vector 60 corresponding to the data variance, rather than predicting each value of the raw data 56. The first principal component of the raw data is used to determine the orientation of the raw data 56 and vector 60. In addition to orientation 62, another parameter, referred to as the orientation of vector 60, is determined, which is indicated as theta (θ). The orientation θ enables an accurate fit of the vector 60 to be obtained covering the variance of the data. The orientation θ is the angle between a horizontal line 64 extending from the origin of [ mu ] and the orientation 62 of the vector 60.

For accuracy, the orientation θ must reflect the correct alignment of vector 60 with the data. Because the orientation θ is initially unknown, the density module 66 is employed to determine the orientation. In the density module 66, the determination of the optimal value for orientation θ is driven by this data.

With respect to fig. 2 and 6, in a density module 66, the data is transformed to polar coordinates 68 and a density versus orientation theta plot 72 is generated. The density is indicated by rho (ρ). Because the data is symmetrical, positive values may be used for simplicity. Furthermore, because the data is not continuous, the first ten percent (10%) of the standard deviation of the data is used to select the data range 70. The density ρ of the data range 70 is calculated based on the median value of the data:

the median value of the orientation θ is used to determine the density center of the data range 70, which corresponds to the optimal value of the orientation.

With further regard to fig. 7, once the orientation θ is determined, the data range 70 is transformed back to the μ versus glide 46 plot and the polar coordinates are used to calculate the orientation 62 of the vector 60. Specifically, the coefficient of the second polar axis PC2 is divided by the coefficient of the first polar axis PC1 to determine the orientation 62. In this manner, the orientation 62 and orientation θ of the vector 60 are determined, thereby denoising the data 74.

Once the data has been denoised 74 by determining the orientation 62 and orientation θ of the vector 60, the stiffness calculator 76 ascertains the slope of the vector. The slope of vector 60 is for the longitudinal stiffness estimate 52 of the tire 12. Preferably, the longitudinal stiffness estimate 52 is communicated to other vehicle control systems through the CAN bus system 28 for use in such systems and/or for determining certain conditions of the tire 12.

For example, turning to FIG. 8, the longitudinal stiffness estimate 52 may be used in a road condition monitor 106 to monitor the condition of the road surface on which the tire 12 is traveling by differentiating between different road surface conditions 88. The stiffness 52 of the tire 12 exhibits rapid time-varying characteristics over different road conditions 88. More specifically, in the non-temperature compensated plot 78, each tire stiffness estimate 52 is different for the summer tire 80, the all season tire 82, and the winter tire 84 on the dry surface 90, the wet surface 92, the snow covered surface 94, and the ice surface 96, respectively.

Because tire stiffness is temperature sensitive, it is important to correct the tire stiffness estimate 52 for the effects of temperature, as shown in the temperature compensation plot 86. From the temperature compensation plot 86, it can be seen that the temperature compensation may exaggerate the differences in the stiffness estimates 52, particularly for the summer tire 80 and the full season tire 82. Further, it can be seen that the stiffness estimate 52 for the winter tire 84 generally exhibits a lower dependence on the type of road surface 88, while the stiffness estimates for the summer tire 80 and the full season tire 82 are higher on the ice surface 96 than on the snow covered surface 94. Based on this information, the longitudinal stiffness estimate 52 may therefore be used by the road monitor 106 to distinguish between the dry road 90, the wet road 92, the snow covered road 94, and the icy road 96.

With respect to fig. 9, the longitudinal stiffness estimate 52 may be used in the wear state monitor 108 to monitor the wear state of the tire 12 by differentiating between different wear states. The stiffness 52 of the tire 12 exhibits a slow time-varying characteristic as it wears. More specifically, in the μ versus slip plot 98 for the worn tire 100, the stiffness estimate 52 is at least thirty percent (30%) higher than the stiffness estimate in the μ versus slip plot 102 for the worn tire 104. Based on this information, the longitudinal stiffness estimate 52 may therefore be used by the wear state monitor 108 to distinguish between different wear states of the tire 12.

In this manner, the longitudinal stiffness estimation system 10 of the present invention provides the stiffness estimate 52 for the tire 12 in real-time based on input signals from a standard vehicle system, such as the CAN bus system 28. Accordingly, the longitudinal stiffness estimation system 10 of the present invention provides an accurate stiffness estimate 52 based on a minimum of data sources. Furthermore, the use of the denoising module 58 described above in the longitudinal stiffness estimation system 10 involves low computational load and thus can be performed on the vehicle-based processor 26, as opposed to prior art systems that involve high computational load and must be performed remotely.

The present invention also includes a method of estimating the longitudinal stiffness of a tire 12. The method comprises the steps according to the description presented above and shown in fig. 1 to 9. The longitudinal stiffness estimation system 10 and accompanying method of the present invention may be referred to as the Nouri technique.

It is to be understood that the structure and method of stiffness estimation system 10 described above may be altered or rearranged, or components or steps known to those skilled in the art omitted or added, without affecting the overall concept or operation of the invention. For example, while the system 10 is described above using acceleration conditions of the vehicle 14, the system may be applied to cruise and brake conditions without affecting the overall concept or operation of the present invention.

Example embodiment

Example 1 includes a longitudinal stiffness estimation system for at least one tire of a supporting vehicle, the longitudinal stiffness estimation system comprising: an electronic communication system disposed on the vehicle; at least one sensor disposed on the vehicle and in electronic communication with the electronic communication system; a processor accessible through the electronic communication system; the at least one sensor measures a selected parameter associated with the vehicle and communicates data of the selected parameter to the processor through the electronic communication system; a mu slip curve generator in communication with the processor and receiving the selected parameters and generating a mu slip curve in real time from the transmitted data; an extraction module in communication with the processor and extracting raw data from a linear portion of the mu slip curve; a denoising module in communication with the processor and denoising the raw data from the mu slip curve by determining from the raw data a pattern represented by a vector, an orientation of the vector, and an orientation of the vector, wherein the denoising module generates denoised data; and a stiffness calculator that receives the denoising data and generates a longitudinal stiffness estimate for the at least one tire.

Example 2 includes the longitudinal stiffness estimation system of example 1, wherein the selected parameter includes at least one of a longitudinal acceleration of the vehicle, a wheel speed, and a vehicle reference speed.

Example 3 includes the longitudinal stiffness estimation system of example 2, wherein the mu slip curve generator receives a longitudinal acceleration of the vehicle to approximate mu in the mu slip curve as a coefficient of friction between a tire and a surface over which the tire is traveling.

Example 4 includes the longitudinal stiffness estimation system of example 2, wherein the mu slip profile generator receives a wheel speed and a vehicle reference speed to calculate slip in the mu slip profile.

Example 5 includes the longitudinal stiffness estimation system of example 1, further comprising an acceleration module that receives a throttle position and a brake pedal position to determine whether the vehicle is accelerating.

Example 6 includes the longitudinal stiffness estimation system of example 1, wherein the pattern is determined as a feature vector of the raw data.

Example 7 includes the longitudinal stiffness estimation system of example 1, wherein the orientation of the vector is an angle between a horizontal line extending from an origin of mu in the mu slip curve and an orientation of the vector.

Example 8 includes the longitudinal stiffness estimation system of example 1, wherein the denoising module includes a density module to determine an orientation of the vector, and the density module transforms the raw data into polar coordinates and generates a density versus orientation plot.

Example 9 includes the longitudinal stiffness estimation system of example 8, wherein the density module determines the selected data range in the density versus orientation plot and calculates the density of the selected data range based on a median value of the selected data range.

Example 10 includes the longitudinal stiffness estimation system of example 9, wherein the density module determines a density center of the selected data range, the density center corresponding to an optimal value of the orientation of the vector.

Example 11 includes the longitudinal stiffness estimation system of example 1, wherein the denoising module uses polar coordinates to determine an orientation of the vector, wherein coefficients of a second polar axis are divided by coefficients of a first polar axis to determine the orientation of the vector.

Example 12 includes the longitudinal stiffness estimation system of example 1, wherein the stiffness calculator ascertains a slope of the vector to generate the longitudinal stiffness estimate.

Example 13 includes the longitudinal stiffness estimation system of example 1, wherein the processor is mounted on the vehicle.

Example 14 includes the longitudinal stiffness estimation system of example 1, wherein the processor is a cloud computing processor.

Example 15 includes the longitudinal stiffness estimation system of example 1, wherein the system provides a longitudinal stiffness estimate for each tire mounted on a drive wheel of the vehicle.

Example 16 includes the longitudinal stiffness estimation system of example 1, wherein the longitudinal stiffness estimate is communicated to a vehicle control system via the electronic communication system.

Example 17 includes the longitudinal stiffness estimation system of example 1, further comprising a road surface condition monitor, wherein the road surface condition monitor receives a plurality of longitudinal stiffness estimates and based on time-varying characteristics of the longitudinal stiffness estimates, the road surface condition monitor distinguishes between at least two of a dry road surface, a wet road surface, a snow-covered road surface, and an icy road surface.

Example 18 includes the longitudinal stiffness estimation system of example 17, wherein the road surface condition monitor includes temperature correction for each longitudinal stiffness estimate.

Example 19 includes the longitudinal stiffness estimation system of example 1, further comprising a wear state monitor, wherein the wear state monitor receives a plurality of longitudinal stiffness estimates and based on time-varying characteristics of the longitudinal stiffness estimates, the wear state monitor distinguishes between different wear states of the tire.

Example 20 includes the longitudinal stiffness estimation system of example 19, wherein the wear state monitor is to distinguish between different wear states of the tire based on a worn tire stiffness estimate that is higher than a new tire stiffness estimate.

The invention has been described with respect to preferred embodiments. Modifications and alterations will occur to others upon a reading and understanding of this specification. It is to be understood that all such modifications and alterations are intended to be included within the scope of the invention as set forth in the following claims or the equivalents thereof.

Claims

1. A longitudinal stiffness estimation system for at least one tire of a supporting vehicle, the longitudinal stiffness estimation system characterized by:

an electronic communication system disposed on the vehicle;

at least one sensor disposed on the vehicle and in electronic communication with the electronic communication system;

a processor accessible through the electronic communication system;

the at least one sensor measures a selected parameter associated with the vehicle and communicates data of the selected parameter to the processor through the electronic communication system;

a mu slip curve generator in communication with the processor and receiving the selected parameters and generating a mu slip curve in real time from the transmitted data;

an extraction module in communication with the processor and extracting raw data from a linear portion of the mu slip curve;

a denoising module in communication with the processor and denoising the raw data from the mu slip curve by determining from the raw data a pattern represented by a vector, an orientation of the vector, and an orientation of the vector, characterized in that the denoising module generates denoising data; and

a stiffness calculator that receives the denoising data and generates a longitudinal stiffness estimate for the at least one tire.

2. The longitudinal stiffness estimation system of claim 1, wherein the selected parameter comprises at least one of a longitudinal acceleration of the vehicle, a wheel speed, and a vehicle reference speed.

3. The longitudinal stiffness estimation system of claim 2 wherein the mu slip curve generator receives longitudinal acceleration of the vehicle to approximate mu in the mu slip curve as a coefficient of friction between a tire and a surface over which the tire is traveling.

4. The longitudinal stiffness estimation system of claim 2 wherein the mu slip curve generator receives wheel speeds and vehicle reference speeds to calculate slip in the mu slip curve.

5. The longitudinal stiffness estimation system of claim 1 further characterized by an acceleration module that receives a throttle position and a brake pedal position to determine whether the vehicle is accelerating.

6. The longitudinal stiffness estimation system of claim 1 wherein the pattern is determined as a feature vector of the raw data.

7. The longitudinal stiffness estimation system of claim 1 wherein the orientation of the vector is an angle between a horizontal line extending from an origin of mu in the mu slip curve and the orientation of the vector.

8. The longitudinal stiffness estimation system of claim 1 wherein the denoising module comprises a density module to determine an orientation of the vector, and the density module transforms the raw data into polar coordinates and generates a density versus orientation plot.

9. The longitudinal stiffness estimation system of claim 8 wherein the density module determines the selected data range in the density versus orientation plot and calculates the density of the selected data range based on a median value of the selected data range.

10. The longitudinal stiffness estimation system of claim 9 wherein the density module determines a density center of the selected data range, the density center corresponding to an optimal value of the orientation of the vector.