Method and system for detecting swapped tires on a vehicle
AssigneeThe Goodyear Tire & Rubber Company
InventorsDario Torregrossa, Oliviero Trufffo, Peter Jung-min Suh
Life of this patent
Estimating the term…
Abstract
The invention provides a method and system for automatically detecting a swap of tires on a vehicle using RSSI levels obtained from tires equipped with transmitters. The obtained RSSI level distributions are used to artificially generate a set of training records, allowing for robust training of a machine learning algorithm. The machine learning algorithm is thereby enabled to provide efficient detection of tire swaps, and to generate a corresponding output signal.
Claims20
A computer-implemented method for detecting a swap of tires on a vehicle including a set of tires mounted in a plurality of available mounting positions of the vehicle, each tire being equipped with a signal transmitter, the method comprising the steps of:
- recording a set of initial mounting positions of the tires in a first memory element of a control unit of the vehicle;
- providing initial signal signature data in a second memory element of the control unit, wherein the initial signal signature data is obtained by receiving, at a signal receiver of the vehicle, signals from the signal transmitters while the tires are mounted at their respective initial mounting positions;
- training a machine learning algorithm of the control unit using a training data set, so that the trained machine learning algorithm is enabled to classify input signal signature data, obtained by receiving signals from the signal transmitters of the tires, as either corresponding to the initial signal signature data, or not;
- obtaining input signal signature data at the signal receiver of the vehicle and classifying the input signal signature data using the trained machine learning algorithm;
- generating a tire swap detection signal at the control unit if the input signal signature data is classified as not corresponding to the initial signal signature data;
- wherein the training data set comprises:
- matched records, in which the initial signal signature data is associated with the set of corresponding initial mounting positions, and
- mismatched records, which are generated by a processor by associating the initial signal signature data with a plurality of sets of mounting positions that differ from the set of initial mounting positions.
A tire swap detection system comprising a control unit and a set of tires mounted in a plurality of available mounting positions of a vehicle, each tire being equipped with a signal transmitter, wherein the control unit comprises a signal receiver for receiving signals from the signal transmitters, a first memory element for storing initial mounting positions of the tires and a second memory element for storing initial signal signature data, the initial signal signature data having been obtained by receiving, at the signal receiver, signals from the signal transmitters while the tires are mounted at their respective initial mounting positions, wherein the control unit further comprises a trained machine learning algorithm configured to classify input signal signature data, obtained by receiving signals from the signal transmitters of the tires, as either corresponding to the initial signal signature data or not, and wherein the control unit is further configured to generate a tire swap detection signal if input signature data is classified as not corresponding to the initial signal signature data.
A control unit for a vehicle, configured to generate a tire swap detection signal, wherein the control unit comprises a signal receiver for receiving signals from signal transmitters of tires, a first memory element for storing initial mounting positions of the tires on the vehicle and a second memory element for storing initial signal signature data, the initial signal signature data having been obtained by receiving, at the signal receiver, signals from the signal transmitters while the tires are mounted at their respective initial mounting positions on the vehicle, wherein the control unit further comprises a trained machine learning algorithm configured to classify input signal signature data, obtained by receiving signals from the signal transmitters of the tires, as either corresponding to the initial signal signature data or not, and wherein the control unit is further configured to generate a tire swap detection signal if input signature data is classified as not corresponding to the initial signal signature data.
Dependencies are inferred from the claim text.