IP Library › Granted Patent US 12,370,912

Granted Patent B1

US 12,370,912 · App. 19/089,649 · Granted Jul 29, 2025

Systems and methods for detecting vehicle position during dynamic wireless charging

Inventors: Arif I. Sarwat (Miami, FL); Milad Behnamfar (Miami, FL); Mohd Tariq (Miami, FL)

Assignee: The Florida International University Board of Trustees

B60L53/126 G06N20/20 B60L2240/12 B60L2240/62

View Patent ↗

Loading inventors, assignments & file history…

Monitor This Case

Get email alerts when status or documents change.

Order Certified Copies

Most orders are placed with the USPTO same day — all within 24 business hours.

Order via The Patent Place →

Pre-filled with this patent's details

Quick Facts

Patent No.: US 12,370,912
App. No.: 19/089,649
Granted: Jul 29, 2025
Kind: B1

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

Systems and methods are provided for detecting vehicle position during electric vehicle charging using machine learning, as well as a control scheme for activating charging coils when the vehicle crosses them. The machine learning can have as inputs the primary current, the vehicle ground clearances, and the speed of the vehicle. As different vehicles with different ground clearances are traveling across the transmitter coils for charging, ground clearance can be defined as a variable for machine learning. Vehicles can move at different speeds, so the speed of the vehicle can be considered in the prediction of the vehicle position.

Claims (56)

1. A system for detecting position of a vehicle during wireless charging, the system comprising:

at least two transmitter coils configured to perform wireless charging with a receiver coil;

at least one first sensor configured to sense a ground clearance of the vehicle during wireless charging;

at least one second sensor configured to sense a speed of the vehicle during wireless charging; and

a controller in operable communication with the at least two transmitter coils, the at least one first sensor, and the at least one second sensor,

the controller being configured to perform the following steps during wireless charging:

i) receive, from the at least two transmitter coils during wireless charging, a primary current of the at least two transmitter coils;

ii) receive, from the at least one first sensor during wireless charging, a measurement of the ground clearance of the vehicle;

iii) receive, from the at least one second sensor during wireless charging, a measurement of the speed of the vehicle;

iv) utilize a machine learning algorithm to determine a position of the vehicle in real time based on the primary current of the at least two transmitter coils, the ground clearance of the vehicle, and the speed of the vehicle; and

v) implement a control strategy to control the at least two transmitter coils based on the position of the vehicle.

2. The system according to claim 1 , the control strategy comprising controlling respective switches of the at least two transmitter coils to ensure a transmitter coil of the at least two transmitter coils that is closest to the receiver coil is activated and all other transmitter coils of the at least two transmitter coils are deactivated.

3. The system according to claim 1 , the controller being configured to further perform the following step during wireless charging:

vi) repeat steps i)-v) until the controller determines that the receiver coil is no longer within range to be wirelessly charged by any transmitter coil of the at least two transmitter coils.

4. The system according to claim 1 , the machine learning algorithm using only the primary current of the at least two transmitter coils, the ground clearance of the vehicle, and the speed of the vehicle as inputs, and generating the position of the vehicle as output.

5. The system according to claim 1 , the machine learning algorithm being random forest, decision tree, gradient boosting, K-nearest neighbor (KNN), support vector regression (SVR), neural network, Bayesian ridge, or linear regression.

6. The system according to claim 1 , the machine learning algorithm being random forest or decision tree.

7. The system according to claim 1 , the machine learning algorithm being random forest.

8. The system according to claim 1 , the at least two transmitter coils comprising five transmitter coils.

9. The system according to claim 1 , the at least two transmitter coils being directly connected to at least two compensation networks, respectively.

10. The system according to claim 9 , the at least two compensation networks being directly connected to at least two inverters, respectively.

11. A method for detecting position of a vehicle during wireless charging, the method comprising:

i) providing at least two transmitter coils configured to perform wireless charging with a receiver coil, at least one first sensor configured to sense a ground clearance of the vehicle during wireless charging, and at least one second sensor configured to sense a speed of the vehicle during wireless charging;

ii) performing wireless charging of the vehicle using the at least two transmitter coils;

iii) receiving, by a controller, from the at least two transmitter coils during wireless charging, a primary current of the at least two transmitter coils;

iv) receiving, by the controller, from the at least one first sensor during wireless charging, a measurement of the ground clearance of the vehicle;

v) receiving, by the controller, from the at least one second sensor during wireless charging, a measurement of the speed of the vehicle;

vi) utilizing, by the controller, a machine learning algorithm to determine a position of the vehicle in real time based on the primary current of the at least two transmitter coils, the ground clearance of the vehicle, and the speed of the vehicle; and

vii) implementing, by the controller, a control strategy to control the at least two transmitter coils based on the position of the vehicle.

12. The method according to claim 11 , the control strategy comprising controlling respective switches of the at least two transmitter coils to ensure a transmitter coil of the at least two transmitter coils that is closest to the receiver coil is activated and all other transmitter coils of the at least two transmitter coils are deactivated.

13. The method according to claim 11 , further comprising:

viii) repeating steps iii)-vii), by the controller, until the controller determines that the receiver coil is no longer within range to be wirelessly charged by any transmitter coil of the at least two transmitter coils.

14. The method according to claim 11 , the machine learning algorithm using only the primary current of the at least two transmitter coils, the ground clearance of the vehicle, and the speed of the vehicle as inputs, and generating the position of the vehicle as output.

15. The method according to claim 11 , the machine learning algorithm being random forest, decision tree, gradient boosting, K-nearest neighbor (KNN), support vector regression (SVR), neural network, Bayesian ridge, or linear regression.

16. The method according to claim 11 , the machine learning algorithm being random forest or decision tree.

17. The method according to claim 11 , the machine learning algorithm being random forest.

18. The method according to claim 11 , the at least two transmitter coils comprising five transmitter coils.

19. The method according to claim 11 , the at least two transmitter coils being directly connected to at least two compensation networks, respectively, and

the at least two compensation networks being directly connected to at least two inverters, respectively.

20. A system for detecting position of a vehicle during wireless charging, the system comprising: