Method for determining road surface condition

• US 10,525,777 B2
• Filed: 11/17/2015
• Issued: 01/07/2020
• Est. Priority Date: 12/05/2014
• Status: Active Grant

• Current Assignee: Bridgestone Corporation
• Original Assignee: Bridgestone Corporation
• Inventors: Goto, Takato, Hanatsuka, Yasushi
• Primary Examiner(s): Sinha, Tarun
• Application Number:
  US15/532,572
• Publication Number:
  US 20180264894A1
• Time in Patent Office: 1,512 Days
• Field of Search: 73584
• CPC Class Codes: B60C 11/246 Tread wear monitoring systems
• CPC Class Codes: B60C 19/00 Tyre parts or constructions not otherwise provided for
• CPC Class Codes: B60C 2019/004 Tyre sensors other than for detecting tyre pressure
• CPC Class Codes: B60C 23/064 comprising tyre mounted deformation sensors, e.g. to determine road contact area
• CPC Class Codes: B60T 2210/12 Friction
• CPC Class Codes: B60T 8/172 Determining control parameters used in the regulation, e.g. by calculations involving measured or detected parameters B60T8/17551 takes precedence
• CPC Class Codes: B60W 2552/00 Input parameters relating to infrastructure
• CPC Class Codes: B60W 40/06 Road conditions
• CPC Class Codes: B60W 40/064 Degree of grip
• CPC Class Codes: B60W 40/068 Road friction coefficient

Abstract

In a method for determining a road surface condition, the speed of calculation is improved by applying an appropriate path limitation. To further improve the accuracy of determination, a time-series waveform of tire vibration detected by an acceleration sensor is windowed at time width T by a windowing means, and time-series waveforms of tire vibration in the respective time windows are extracted before feature vectors X_i of the respective time windows are calculated. Then kernel functions K_A (X, Y) are calculated from the feature vectors X_i for the respective time windows and road surface feature vectors Y_i, which are feature vectors for the respective time windows calculated from time-series waveforms of tire vibration having been determined in advance for distinctive road surface conditions. In doing so, the kernel functions are calculated after weighting respective paths by weighting parameters ω (i, j) using a time-series waveform length corresponding value in the time-series waveform of tire vibration as measured and a previously determined time-series waveform length corresponding value in the time-series waveform of tire vibration.

Claims

1. A method for determining a condition of a road surface under a tire in motion by detecting tire vibration by a vibration detecting means disposed within the tire, the method comprising:

detecting the tire vibration of the tire in motion;

extracting a time-series waveform of the detected tire vibration;

extracting time-series waveforms in respective time windows of a predetermined time width by windowing the time-series waveform of the detected tire vibration;

calculating feature vectors for the time-series waveforms in the respective time windows;

calculating a kernel function from the calculated feature vectors and road surface feature vectors, the road surface feature vectors being feature vectors of the respective time windows calculated from the time-series waveforms of the detected tire vibration having been determined in advance for distinctive road surface conditions; and

determining a road surface condition based on values of discriminant functions using the kernel function;

wherein the kernel function is one of a global alignment kernel function, a dynamic time warping kernel function, or a computed values of the kernel function,

wherein in the calculating the kernel function, the kernel function is calculated after each of path elements calculated from one of the calculated feature vectors for the respective time widths and one of the road surface feature vectors is weighted by a weighting parameter ω

using a first time-series waveform length corresponding value in the time-series waveform of the detected tire vibration and a previously determined second time-series waveform length corresponding value in another time-series waveform of tire vibration, and

wherein in the determining the road surface condition, the road surface condition is determined by comparing the values of discriminant functions determined for the distinctive road surface conditions.

2. The method for determining a road surface condition according to claim 1, wherein the first time-series waveform length corresponding value is any one or a plurality of the length of the time-series waveform of the detected tire vibration, the tire rotation speed at the acquisition of the time-series waveform of the detected tire vibration, and the inter-peak distance in the time-series waveform of the detected tire vibration, or the computed value of any one or a plurality of the length of the time-series waveform of the detected tire vibration, the tire rotation speed at the acquisition of the time-series waveform of the detected tire vibration, and the inter-peak distance in the time-series waveform of the detected tire vibration.

3. The method for determining a road surface condition according to claim 1, wherein in the calculating the kernel functions, the paths for calculating kernel functions are only the paths passing through one or both of the path elements corresponding to the feature vectors including the leading-end peak and the road surface feature vectors including the leading-end peak and the path elements corresponding to the feature vectors including the trailing-end peak and the road surface feature vectors including the trailing-end peak.

4. The method for determining a road surface condition according to claim 1, wherein the depth of tire wear is added to the components of the feature vectors.

5. The method for determining a road surface condition according to claim 1, wherein the components of the feature vectors to be employed are the computed values of the feature vectors calculated respectively from the time-series waveforms of the respective time windows and the depths of tire wear.

6. The method for determining a road surface condition according to claim 1, wherein the coefficient of the discriminant functions is the function of the depth of tire wear.