Granted Patent B2

US 12,304,267 · App. 18/044,851 · Granted May 20, 2025

Inventor: Chidambaram Subramanian (Greensboro, NC)

Assignee: VOLVO TRUCK CORPORATION

B60G17/0195B60C23/002B60C99/006B60G17/0165B60G17/052B60C2019/004B60G2400/204B60G2400/52B60G2400/82B60G2500/201

Abstract

Claims (110)

1. A method, comprising:

receiving, at a processor from a plurality of sensors, a plurality of vehicle information associated with a vehicle, the plurality of vehicle information comprising:

a vehicle velocity;

at least one tire pressure;

a normal load;

a road surface; and

an inertial movement unit at a center of gravity of the vehicle;

executing, via the processor, a first machine learning algorithm, wherein:

first machine learning algorithm inputs comprise the plurality of vehicle information; and

first machine learning algorithm outputs comprise a lateral dimension, a longitudinal dimension, and a diagonal dimension, the lateral dimension, the longitudinal dimension, and the diagonal dimension defining a tire contact patch of at least one tire on the vehicle;

receiving, at the processor, a human preference for how the vehicle operates; and

executing, via the processor, a second machine learning algorithm, wherein:

second machine learning algorithm inputs comprise the plurality of vehicle information, the first machine learning algorithm outputs, and the human preference; and

second machine learning algorithm outputs comprise a desired tire pressure of the at least one tire and an air suspension adjustment for the normal load of the vehicle.

2. The method of claim 1 , further comprising:

modifying the at least one tire pressure to the desired tire pressure; and

modifying air suspension of the vehicle according to the air suspension adjustment.

3. The method of claim 2 , wherein:

the modifying of the at least one tire pressure occurs via a tire pressure actuator; and

the modifying of the air suspension occurs via an air suspension actuator.

4. The method of claim 1 , further comprising:

training a first neural network by:

generating a first sensitivity analysis for the plurality of vehicle information and known tire contact patches; and

forming a first plurality of nodes and a first plurality of neural layers linking the first plurality of nodes based on the first sensitivity analysis, resulting in a first trained neural network; and

converting the first trained neural network to executable code, resulting in the first machine learning algorithm.

5. The method of claim 4 , further comprising:

training a second neural network by:

generating a second sensitivity analysis for the plurality of vehicle information, the first machine learning algorithm outputs, known tire pressures, and known air suspension adjustments for the normal load; and

forming a second plurality of nodes and a second plurality of neural layers linking the second plurality of nodes based on the second sensitivity analysis, resulting in a second trained neural network; and

converting the second trained neural network to executable code, resulting in the second machine learning algorithm.

6. The method of claim 1 , wherein the human preference comprises at least one of:

maximum fuel economy;

minimize tire wear;

maximum ride comfort; and

maximum performance.

7. The method of claim 1 , wherein both the first machine learning algorithm and the second machine learning algorithm identify multi-dimensional boundary conditions associated with respective outputs.

8. A vehicle comprising:

a plurality of sensors;

a processor; and

a non-transitory computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:

receiving, from the plurality of sensors, a plurality of vehicle information associated with the vehicle, the plurality of vehicle information comprising:

a vehicle velocity;

at least one tire pressure;

a normal load;

a road surface; and

an inertial movement unit at a center of gravity of the vehicle;

executing a first machine learning algorithm, wherein:

first machine learning algorithm inputs comprise the plurality of vehicle information; and

first machine learning algorithm outputs comprise a lateral dimension, a longitudinal dimension, and a diagonal dimension, the lateral dimension, the longitudinal dimension, and the diagonal dimension defining a tire contact patch of at least one tire on the vehicle;

receiving a human preference for how the vehicle operates; and

executing a second machine learning algorithm, wherein:

second machine learning algorithm inputs comprise the plurality of vehicle information, the first machine learning algorithm outputs, and the human preference; and

second machine learning algorithm outputs comprise a desired tire pressure of the at least one tire and an air suspension adjustment for the normal load of the vehicle.

9. The vehicle of claim 8 , the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:

modifying the at least one tire pressure to the desired tire pressure; and

modifying air suspension of the vehicle according to the air suspension adjustment.

10. The vehicle of claim 9 , wherein:

the modifying of the at least one tire pressure occurs via a tire pressure actuator; and

the modifying of the air suspension occurs via an air suspension actuator.

11. The vehicle of claim 8 , the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:

training a first neural network by:

generating a first sensitivity analysis for the plurality of vehicle information and known tire contact patches; and

forming a first plurality of nodes and a first plurality of neural layers linking the first plurality of nodes based on the first sensitivity analysis, resulting in a first trained neural network; and

converting the first trained neural network to executable code, resulting in the first machine learning algorithm.

12. The vehicle of claim 11 , the non-transitory computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:

training a second neural network by:

generating a second sensitivity analysis for the plurality of vehicle information, the first machine learning algorithm outputs, known tire pressures, and known air suspension adjustments for the normal load;

forming a second plurality of nodes and a second plurality of neural layers linking the second plurality of nodes based on the second sensitivity analysis, resulting in a second trained neural network; and

converting the second trained neural network to executable code, resulting in the second machine learning algorithm.

13. The vehicle of claim 8 , wherein the human preference comprises at least one of:

maximum fuel economy;

minimize tire wear;

maximum ride comfort; and

maximum performance.

14. The vehicle of claim 8 , wherein both the first machine learning algorithm and the second machine learning algorithm identify multi-dimensional boundary conditions associated with respective outputs.

15. A non-transitory computer-readable storage medium having instructions stored which, when executed by a processor, cause the processor to perform operations comprising:

receiving, from a plurality of sensors, a plurality of vehicle information associated with a vehicle, the plurality of vehicle information comprising:

a vehicle velocity;

at least one tire pressure;

a normal load;

a road surface; and

an inertial movement unit at a center of gravity of the vehicle;

executing a first machine learning algorithm, wherein:

first machine learning algorithm inputs comprise the plurality of vehicle information; and

first machine learning algorithm outputs comprise a lateral dimension, a longitudinal dimension, and a diagonal dimension, the lateral dimension, the longitudinal dimension, and the diagonal dimension defining a tire contact patch of at least one tire on the vehicle;

receiving a human preference for how the vehicle operates; and

executing a second machine learning algorithm, wherein:

second machine learning algorithm inputs comprise the plurality of vehicle information, the first machine learning algorithm outputs, and the human preference; and

second machine learning algorithm outputs comprise a desired tire pressure of the at least one tire and an air suspension adjustment for the normal load of the vehicle.

16. The non-transitory computer-readable storage medium of claim 15 , having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:

modifying the at least one tire pressure to the desired tire pressure; and

modifying air suspension of the vehicle according to the air suspension adjustment.

17. The non-transitory computer-readable storage medium of claim 16 , wherein:

the modifying of the at least one tire pressure occurs via a tire pressure actuator; and

the modifying of the air suspension occurs via an air suspension actuator.

18. The non-transitory computer-readable storage medium of claim 15 , having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:

training a first neural network by:

generating a first sensitivity analysis for the plurality of vehicle information and known tire contact patches; and

forming a first plurality of nodes and a first plurality of neural layers linking the first plurality of nodes based on the first sensitivity analysis, resulting in a first trained neural network; and

converting the first trained neural network to executable code, resulting in the first machine learning algorithm.

19. The non-transitory computer-readable storage medium of claim 18 , having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising:

training a second neural network by:

generating a second sensitivity analysis for the plurality of vehicle information, the first machine learning algorithm outputs, known tire pressures, and known air suspension adjustments for the normal load;

forming a second plurality of nodes and a second plurality of neural layers linking the second plurality of nodes based on the second sensitivity analysis, resulting in a second trained neural network; and

converting the second trained neural network to executable code, resulting in the second machine learning algorithm.

20. The non-transitory computer-readable storage medium of claim 15 , wherein the human preference comprises at least one of:

maximum fuel economy;

minimize tire wear;

maximum ride comfort; and

maximum performance.

Assignments (1)

Continuity (1)

References Cited (14)