Method for operating a distance sensor of a vehicle in which a transmission signal is adapted in accordance with how an object is classified, computing device, and sensor device

Patent metadata

Field	Value
Publication number	US-12019157-B2
Application number	US-202017774702-A
Country	US
Kind code	B2
Filing date	Oct 21, 2020
Priority date	Nov 11, 2019
Publication date	Jun 25, 2024
Grant date	Jun 25, 2024

Abstract

The invention relates to a method for operating a distance sensor (4) of a vehicle (1), in which method a plurality of successive measurement cycles are carried out in an operating mode, wherein, in each measurement cycle, a transmission signal is transmitted, a reception signal (Rx1 to Rx8) is determined on the basis of the transmission signal reflected in a surrounding region (9) of the vehicle (1), the object (8) is classified, and the transmission signal is selected from a plurality of predefined transmission signals in accordance with how the object (8) is classified, wherein the transmission signal is selected in accordance with an assignment rule determined in a learning mode, said assignment rule describing an assignment of the plurality of predefined transmission signals to classes of objects (8), wherein, in each measurement cycle, the object (8) is classified on the basis of the reception signal (Rx1 to Rx8) and the transmission signal is selected in accordance with how the object (8) is classified for subsequent measurement cycles.

First claim

The invention claimed is:

1. A method for operating a distance sensor of a vehicle, comprising:
carrying out a plurality of successive measurement cycles in an operating mode, wherein in each measurement cycle a transmission signal is transmitted, the transmission signal reflected in an area surrounding the vehicle is taken as a basis for determining a received signal; and
classifying at least one object and selecting the transmission signal from a plurality of predetermined transmission signals on the basis of the classification of the at least one object,
the selection of the transmission signal being carried out on the basis of an association rule, determined in a learning mode, that describes an association of the plurality of predetermined transmission signals with classes of objects,
wherein in each measurement cycle the at least one object is classified on the basis of the received signal and the transmission signal is selected for a subsequent measurement cycle on the basis of the classification of the at least one object,
determining whether each of a plurality of objects is classified during the classification of the at least one object,
wherein the plurality of objects comprises the at least one object, and
wherein selecting the transmission signal is based, at least in part, on a prioritization of object classes, in response to the determination that each of the plurality of objects is classified.

2. The method as claimed in claim 1 , wherein in the learning mode, reference measurements are carried out on reference objects, the plurality of predetermined transmission signals being transmitted in the reference measurements using the distance sensor and the respective received signals being determined for the reference objects.

3. The method as claimed in claim 2 , wherein for the association rule is determined on the basis of the respective received signals by a machine learning method.

4. The method as claimed in claim 1 , wherein a reference transmission signal is transmitted as the transmission signal at the start of the operating mode.

5. The method as claimed in claim 1 , wherein if the object is not identified in a measurement cycle, a reference transmission signal is selected in the subsequent measurement cycle.

6. The method as claimed in claim 1 , wherein an amplitude-modulated transmission signal is selected as the reference transmission signal.

7. The method as claimed in claim 1 , wherein if the object is classified as a tall object, an amplitude-modulated transmission signal is selected as the transmission signal in the subsequent measurement cycle.

8. The method as claimed in claim 1 , wherein if the object is classified as a living being, a frequency-modulated transmission signal is selected as the transmission signal in the subsequent measurement cycle.

9. The method as claimed in claim 1 , wherein if at least two objects are identified in the surrounding area, the transmission signal is selected on the basis of the object that is at the shortest distance from the distance sensor.

10. The method as claimed in claim 1 , wherein the plurality of transmission signals differ from one another in terms of a modulation, a frequency and/or a transmission duration.

11. A computing device for a sensor apparatus of a vehicle, wherein the computing device is configured to carry out a method as claimed in claim 1 .

12. A sensor apparatus for a vehicle comprising: a computing device as claimed in claim 11 ; and at least one distance sensor in the form of an ultrasonic sensor.

13. A computer program, comprising instructions that, when the program is executed by a computing device, cause the latter to carry out a method as claimed in claim 1 .

14. A non-transitory computer-readable storage medium, comprising instructions that, when executed by a computing device, cause the latter to carry out a method as claimed in claim 1 .

Assignees

• Valeo Schalter & Sensoren Gmbh

Inventors

• Gotzig Heinrich
• Mohamed Mohamed Elamir

Classifications

• G01S2013/932

  using own vehicle data, e.g. ground speed, steering wheel direction · CPC title

• G01S2007/52007

  involving adjustment of transmitted power · CPC title

• G06N20/00

  Machine learning · CPC title

• G01S7/417

  involving the use of neural networks · CPC title

• G01S7/411

  Identification of targets based on measurements of radar reflectivity (G01S7/415 takes precedence) · CPC title