Computer-Implemented Method for Training an Articial Intelligence Module to Determine a Tire Type of a Motor Vehicle

Abstract

A computer-implemented method for training an artificial intelligence module to determine a tire type of a motor vehicle is disclosed. The method includes providing a measured value dataset on a data carrier, wherein the measured value dataset contains at least one data entry regarding ultrasound data, speed data and tire data, wherein the ultrasound data describe at least one ultrasonic wave that was produced by rolling of a tire of the motor vehicle, wherein the speed data describe a speed of the motor vehicle, wherein the tire data describe a tire type of the motor vehicle. The method further includes generating a modified training dataset based on the measured value dataset. Generating the modified training dataset includes (i) forming an input dataset based on the ultrasound data and the speed data of the measured value dataset, (ii) forming an output dataset based on the tire data of the measured value dataset, and (iii) training the AI module based on the modified training dataset.

Claims

1. A computer-implemented method for training an artificial intelligence module to determine a tire type of a motor vehicle, comprising: providing, on a data carrier, a measured value dataset, wherein the measured value dataset comprises at least one data entry regarding ultrasound data, speed data, and tire data, wherein the ultrasound data describe at least one ultrasonic wave that was produced by rolling of a tire of the motor vehicle, wherein the speed data describe a speed of the motor vehicle, wherein the tire data describe a tire type of the motor vehicle, and generating a modified training dataset based on the measured value dataset, wherein generating the modified training dataset comprises: forming an input dataset based on the ultrasound data and the speed data of the measured value dataset, forming an output dataset based on the tire data of the measured value dataset, and training the AI module based on the modified training dataset.

2. The method according to claim 1, wherein generating the modified training dataset further comprises: rejecting data entries in the measured value dataset which exceed and/or fall below a predetermined limit value.

3. The method according to claim 1, wherein generating the modified training dataset further comprises: downsampling/normalizing the ultrasound data.

4. The method according to claim 1, wherein generating the modified training dataset further comprises: standardizing the ultrasound data and/or the speed data of the measured value dataset.

5. The method according to claim 1, wherein generating the modified training dataset further comprises: forming fractions of time-resolved ultrasound data and/or speed data for modeling time series.

6. The method according to claim 5, wherein a duration of the fraction is 1 to 60 seconds.

7. The method according to claim 1, wherein generating the modified training dataset further comprises: comparing the speed data to a predetermined speed limit value, and rejecting data entries of the ultrasound data in the initial training dataset when the speed data, at the same time as the ultrasound data, reach, exceed and/or fall below the predetermined speed limit value.

8. A computer program that, when executed, instructs a processor to carry out steps of the method according to claim 1.

9. A sensor system for a motor vehicle, comprising: at least one ultrasonic sensor, an AI module which has been trained with the method according to claim 1, wherein the ultrasonic sensor is configured to detect an ultrasonic wave that was produced by rolling of a tire of a motor vehicle on a ground surface, wherein the ultrasonic sensor is connected to the AI module, and wherein the AI module is configured to determine a tire type of the tire of the motor vehicle on the basis of the ultrasonic wave detected by the ultrasonic sensor.

10. A motor vehicle, comprising: at least one tire, a sensor system according to claim 9, and/or a computer-readable medium that stores a computer program according to claim 8, wherein the sensor system is configured to determine a type of the at least one tire.

11. The method according to claim 1, wherein generating the modified training dataset further comprises: downsampling/normalizing the ultrasound data to a frequency of approximately 100 Hz.

12. The method according to claim 1, wherein generating the modified training dataset further comprises: standardizing the ultrasound data and/or the speed data of the measured value dataset by removing the average value and/or scaling to unit variance.

13. The method according to claim 1, wherein generating the modified training dataset further comprises: forming fractions of time-resolved ultrasound data and/or speed data for modeling time series of a time window.

14. The sensor system according to claim 9, wherein: the at least one ultrasonic sensor is a parking ultrasonic sensor, and the ultrasonic sensor is connected to the AI module by way of an Internet connection.

Description

EXEMPLARY EMBODIMENTS

[0054] FIG. 1 shows a flow chart illustrating steps of the computer-implemented method according to one embodiment.

[0055] FIG. 2 shows a flow chart illustrating steps of the computer-implemented method according to one embodiment.

[0056] FIG. 3 shows a sensor system according to one embodiment.

[0057] FIG. 4 shows a motor vehicle according to one embodiment.

[0058] FIG. 1 shows a flow chart illustrating steps of the computer-implemented method 100 for training an AI module 204 to determine a tire type of a motor vehicle 300, comprising the steps of: [0059] providing S1, on a data carrier, a measured value dataset, the measured value dataset comprising at least one data entry regarding ultrasound data, speed data, and tire data, the ultrasound data describing at least one ultrasonic wave that was produced by rolling of a tire 202 of the motor vehicle 300, the speed data describing a speed of the motor vehicle 300, the tire data describing a tire type of the motor vehicle 300, [0060] generating S2 a training dataset based on the measured value dataset, the generation of the training dataset comprising the steps of: [0061] forming an input dataset S2a based on the ultrasound data and the speed data of the measured value dataset, [0062] forming an output dataset S2b based on the tire data of the measured value dataset, [0063] training S3 of the AI module 204 based on the training dataset.

[0064] The advantage of this embodiment can be that with the aid of the method 100, already-existing ultrasonic sensors 202 can be used in a motor vehicle 300 in order to determine a type of a tire 302 of the motor vehicle 300.

[0065] Furthermore, the method can be carried out in the sequence as specified by the reference signs, or also in any other sequence. However, it should be particularly noted that steps S2a to S2h can be carried out in any sequence, in particular in any sequence that is technically expedient. In other words, using a dataset, for example consisting of ultrasound signals, speed, tire type, and additional optional information, such as the type of motor vehicle 300 and various environmental factors, such as passing vehicles, obstacles such as construction sites, or weather data and air pressure data, can be used to train an AI module 204. The dataset or the measurement data can be prepared with the method steps 2a to 2h before the model is trained.

[0066] FIG. 2 shows a flow chart illustrating steps of the further method 110. The further method 110 can have steps S1 to S3, as already explained with respect to FIG. 1. Furthermore, in the case of the expanded method 110, in step S2 a plurality of preparation steps for generating the training dataset can be combined. In this regard, step S2 can include steps S2a, forming an input dataset based on the ultrasound data and the speed data of the measured value dataset, and step S2b, forming an output dataset based on the tire data of the measured value dataset. Furthermore, step S2 of the expanded method 110 may include the step of rejecting S2c data entries in the measured value dataset that exceed or fall below a predetermined limit value. In addition, the method 110 has the step of downsampling/normalizing S2d the ultrasound data, in particular to a frequency of approx. 100 Hertz. Furthermore, step S2 of the further method 110 may include the step of standardizing S2e the ultrasound data and/or the speed data of the measured value dataset, in particular by removing the mean value and/or scaling to unit variance. Furthermore, the expanded method 110 may include the step of forming fractions S2f of time-resolved ultrasound data and/or speed data for modeling time series, in particular an alternating time window. Furthermore, the expanded method 110 may have the step of comparing S2g the speed data to a predetermined speed value. Furthermore, the expanded method 110 can include rejecting S2h data entries of the ultrasound data in the measured value dataset if the speed data, at the same time as the ultrasound data, reach, exceed, and/or fall below the predetermined limit value.

[0067] FIG. 3 shows a sensor system 200 which has an ultrasonic sensor 202 and an AI module 204. The AI module 204 can be connected to the ultrasonic sensor 202 by means of a connection, in particular an Internet connection 206. Furthermore, the AI module 204 can be connected to the ultrasonic sensor 202 with any other form of connection, in particular wired or wireless. The ultrasonic sensor 202 can detect an ultrasonic wave or other sound waves which result from the rolling of a tire 302.

[0068] FIG. 4 shows a motor vehicle 300. The motor vehicle 300 has a sensor system 200. The sensor system 200 has an ultrasonic sensor 202, which can in particular be a parking ultrasonic sensor of the motor vehicle 300, and an AI module 204. The AI module 204 can also be located outside the motor vehicle 300, for example on a server. The AI module 204 and the ultrasonic sensor 202 can be connected to one another by means of a connection 206, in particular by means of an Internet or wireless connection technology. Furthermore, the ultrasonic sensor 202 of the motor vehicle 300 can detect an ultrasonic wave or a sound wave resulting from a rolling of the tire 302 of the motor vehicle 300. Furthermore, the motor vehicle 300 can have a computer-readable medium 304 on which the AI module 204 or training datasets or a computer program can be stored. Furthermore, the motor vehicle 300 can have software which operates on the basis of the results of the AI module 204. This software can detect the tire type on the basis of the ultrasound data and the speed data, but the software cannot further improve the AI module 204. Furthermore, such software can be installed on a plurality of engine control devices in order to detect the tire type of motor vehicles 300, in particular independently of the model.

[0069] Additionally, it should be noted that “comprising” and “including” do not exclude other elements, and the indefinite articles “a” or “an” do not exclude a plurality. Furthermore, it should be noted that features that have been described with reference to any of the above embodiments may also be used in combination with other features of other embodiments described above. Reference signs in the claims are not to be considered as limiting.