Method and device for aligning a lens system

1 . A method for ascertaining an alignment of a lens system, comprising the following steps:

a. aligning the lens system according to a provided first alignment;

b. ascertaining a first refracted optical signal, the first refracted optical signal being ascertained by a refraction of a first emitted optical signal at the lens system aligned according to the first alignment;

c. ascertaining a first attribute value, the first attribute value characterizing an attribute of the first refracted optical signal;

d. training a first machine learning system as a function of the first alignment and the ascertained first attribute value, the machine learning system being configured to ascertain an output for an alignment that characterizes the attribute of the alignment; and

e. ascertaining the alignment of the lens system based on the output of the first machine learning system, wherein the first machine learning system includes a polynomial model, which is configured to ascertain the output that characterizes the attribute for an alignment.

2 . The method as recited in claim 1 , wherein the first machine learning system is pretrained in a step in advance of the method.

3 . The method as recited in claim 1 , wherein the ascertaining of the alignment of the lens system based on the output of the first machine learning system includes the following steps:

f. ascertaining a second alignment in such a way that an output of the first machine learning system ascertained for the second alignment lies within a predefinable value range;

g. ascertaining a second refracted optical signal, wherein the second refracted optical signal is ascertained by a refraction of a second emitted optical signal at the lens system aligned according to the second alignment;

h. ascertaining a second attribute value, the second attribute value characterizing the attribute of the second refracted optical signal;

i. when the second attribute value for the second alignment does not lie within a value range predefinable for the second attribute value, repeating steps d, f, g and h, the second alignment being used as an additional first alignment for training the first machine learning system; otherwise, when the second value for the second alignment lies within the value range predefinable for the second attribute value, providing the second alignment as the alignment of the lens system.

4 . The method as recited in claim 3 , wherein the second alignment is ascertained based on an optimization, a constraint of the optimization characterizing at least one limit of the predefinable value range.

5 . The method as recited in claim 1 , wherein the first alignment is provided based on a Bayesian optimization method.

6 . The method as recited in claim 1 , wherein the first alignment is ascertained based on a second machine learning system, the second machine learning system being designed to determine a change in the alignment based on an alignment.

7 . The method as recited in claim 6 , wherein the second machine learning system is trained using a reinforcement learning method.

8 . The method as recited in claim 1 , wherein the lens system is part of an optical sensor.

9 . A device for aligning a lens system, the device being configured to:

a. align the lens system according to a provided first alignment;

b. ascertain a first refracted optical signal, the first refracted optical signal being ascertained by a refraction of a first emitted optical signal at the lens system aligned according to the first alignment;

c. ascertain a first attribute value, the first attribute value characterizing an attribute of the first refracted optical signal;

d. train a first machine learning system as a function of the first alignment and the ascertained first attribute value, the machine learning system being configured to ascertain an output for an alignment that characterizes the attribute of the alignment; and

e. ascertain the alignment of the lens system based on the output of the first machine learning system, wherein the first machine learning system includes a polynomial model, which is configured to ascertain the output that characterizes the attribute for an alignment.

10 . A non-transitory machine-readable memory medium on which is stored a computer program for ascertaining an alignment of a lens system, the computer program, when executed by a processor, causing the processor to perform the following steps:

a. aligning the lens system according to a provided first alignment;

b. ascertaining a first refracted optical signal, the first refracted optical signal being ascertained by a refraction of a first emitted optical signal at the lens system aligned according to the first alignment;

c. ascertaining a first attribute value, the first attribute value characterizing an attribute of the first refracted optical signal;

d. training a first machine learning system as a function of the first alignment and the ascertained first attribute value, the machine learning system being configured to ascertain an output for an alignment that characterizes the attribute of the alignment; and

e. ascertaining the alignment of the lens system based on the output of the first machine learning system, wherein the first machine learning system includes a polynomial model, which is configured to ascertain the output that characterizes the attribute for an alignment.