US 12,354,369 B2
Method for automatically executing a vehicle function, method for evaluating a computer vision method and evaluation circuit for a vehicle
Ruby Moritz, Wolfsburg (DE); Peter Schlicht, Wolfsburg (DE); Fabian Hüger, Wolfenbüttel (DE); and Yasin Bayzidi, Braunschweig (DE)
Assigned to VOLKSWAGEN AKTIENGESELLSCHAFT, Wolfsburg (DE)
Filed by Volkswagen Aktiengesellschaft, Wolfsburg (DE)
Filed on Mar. 23, 2022, as Appl. No. 17/702,535.
Claims priority of application No. 21164745 (EP), filed on Mar. 24, 2021.
Prior Publication US 2022/0309799 A1, Sep. 29, 2022
Int. Cl. G06V 20/58 (2022.01); B60W 40/10 (2012.01); G06T 7/11 (2017.01); G06T 7/194 (2017.01)
CPC G06V 20/58 (2022.01) [B60W 40/10 (2013.01); G06T 7/11 (2017.01); G06T 7/194 (2017.01); B60W 2420/403 (2013.01); B60W 2420/408 (2024.01)] 18 Claims
OG exemplary drawing
 
18. A method for automatically executing a vehicle function of a vehicle based on spatially resolved raw sensor data for environment perception generated by at least one sensor for environment detection of the vehicle, comprising:
receiving spatially resolved raw sensor data generated by the at least one sensor of the vehicle;
processing sensor data which are characteristic of the spatially resolved raw sensor data by a processor, the processor determining at least one region of interest of the sensor data and at least one class for classifying the region of interest;
processing the sensor data by an evaluation circuit based on the determined region of interest and determining at least one local quality parameter which is characteristic for the quality of the sensor data with respect to at least a section of the region of interest; and
executing the vehicle function in dependence on the local quality parameter, wherein
the sensor data are processed by an artificial neural network, the artificial neural network being supplied with the sensor data as input variables and the artificial neural network maps the input variables in dependence of a parameterizable processing chain to output variables, wherein the output variables are characteristic for at least one region of interest of the sensor data and at least one class for classifying the region of interest.