System and method for tire wear prognostics
A storage maintains coefficients that map histogram data elements to tire wear, the coefficients being trained based on a correlation of histogram data to measured tire wear. A processor is programmed to receive a wear data histogram from a vehicle, utilize the coefficients to translate the wear data histogram into a measure of physical tire wear; and send an alert message indicating the estimated tire wear.
1. A system comprising:
a storage that maintains coefficients that map histogram data elements to tire wear, the coefficients being trained based on a correlation of histogram data to measured tire wear; and
a processor programmed to
receive a wear data histogram from a vehicle,
utilize the coefficients to translate the wear data histogram into an estimated measure of physical tire wear, and
send an alert message indicating the estimated physical tire wear.
2. The system of claim 1 , wherein the wear data histogram includes a distribution of power indicative of wear to tires of the vehicle according to one or more variables.
3. The system of claim 2 , wherein the one or more variables include one of engine power output, ambient temperature, or steering angle.
4. The system of claim 2 , wherein the one or more variables include two of engine power output, ambient temperature, or steering angle.
5. The system of claim 2 , wherein the one or more variables include engine power output, ambient temperature, and steering angle.
6. The system of claim 1 , wherein the processor is further programmed to:
request the wear data histogram from the vehicle; and
send the alert message to a mobile device corresponding to the vehicle.
7. The system of claim 1 , wherein the processor is further programmed to:
receive actual wear data corresponding to a plurality of experimental wear data histograms; and
determine the coefficients using machine learning trained using the actual wear data and the corresponding plurality of experimental wear data histograms.
8. The system of claim 1 , wherein the histogram includes a plurality of bins, and the processor is further programmed to:
multiply a value indicated in each of the plurality of bins by a coefficient corresponding to the respective bin to create a translated product for each bin; and
generate the measure of physical tire wear as a sum of the translated products.
9. A method comprising:
utilizing coefficients that map histogram data elements to tire wear, the coefficients being trained based on a correlation of histogram data to measured tire wear, to translate a wear data histogram, including a distribution of power indicative of wear to tires of a vehicle according to one or more variables, into a measure of physical tire wear.
10. The method of claim 9 , further comprising:
sending a first message recommending tire rotation responsive to the measure of physical tire wear exceeding a first wear threshold; and
sending a second message recommending tire replacement responsive to the measure of physical tire wear exceeding a second wear threshold, the second wear threshold indicating a greater level of wear than the first wear threshold.
11. The method of claim 9 , wherein the one or more variables include one of engine power output, ambient temperature, or steering angle.
12. The method of claim 9 , wherein the one or more variables include two of engine power output, ambient temperature, or steering angle.
13. The method of claim 9 , wherein the one or more variables include engine power output, ambient temperature, and steering angle.
14. The method of claim 9 , further comprising determining the coefficients using machine learning trained using a plurality of experimental wear data histograms and actual wear data corresponding to the plurality of experimental wear data histograms.
15. The method of claim 9 , further comprising generating the measure of physical tire wear as a sum of translated products computed by multiplying a value indicated in each of a plurality of bins of the wear data histogram by a coefficient corresponding to the respective bin.
16. A non-transitory computer-readable medium comprising instructions that, when executed by a processor, cause the processor to utilize coefficients that map histogram data elements to tire wear, the coefficients being trained based on a correlation of histogram data to measured tire wear, and to translate a wear data histogram, including a distribution of power indicative of wear to tires of a vehicle according to one or more variables, into a measure of physical tire wear.
17. The medium of claim 16 , further comprising instructions that, when executed by the processor, cause the processor to:
send a first message recommending tire rotation responsive to the measure of physical tire wear exceeding a first wear threshold; and
send a second message recommending tire replacement responsive to the measure of physical tire wear exceeding a second wear threshold, the second wear threshold indicating a greater level of wear than the first wear threshold.
18. The medium of claim 16 , wherein the one or more variables include one of engine power output, ambient temperature, or steering angle.
19. The medium of claim 16 , further comprising instructions that, when executed by the processor, cause the processor to determine the coefficients using machine learning trained using a plurality of experimental wear data histograms and actual wear data corresponding to the plurality of experimental wear data histograms.
20. The medium of claim 16 , further comprising instructions that, when executed by the processor, cause the processor to generate the measure of physical tire wear as a sum of translated products computed by multiplying a value indicated in each of a plurality of bins of the wear data histogram by a coefficient corresponding to the respective bin.