| CPC G06F 16/215 (2019.01) [G06F 16/2365 (2019.01); G06F 16/258 (2019.01); G06F 16/29 (2019.01)] | 14 Claims |
|
1. A method for generating training datasets for training an evaluation algorithm using which an alignment of two map datasets can be evaluated in order to determine navigation information for a mobile device that is moving or is to move in an environment, the method comprising:
for each of a plurality of training datasets:
providing two input feature datasets, which correspond to two map datasets or which have been determined based on the two map datasets, wherein the two map datasets each include environmental information, wherein the environmental information has, in each case, been acquired from the mobile device and/or from the environment using a sensor of the mobile device;
providing a transformation dataset, wherein the transformation dataset has been generated during an alignment of the two input feature datasets, and wherein the transformation dataset includes information about a transformative relation between the two input feature datasets;
providing a reference transformation dataset as ground truth;
determining a correlation dataset based on the two input feature datasets and/or the transformation dataset, wherein the correlation dataset includes information about a correlation between at least two of the following datasets: the two input feature datasets, the transformation dataset;
determining a quality measure depending on an accuracy of a match between the transformation dataset and the reference transformation dataset; and
providing the training dataset, wherein the training dataset includes the correlation dataset and the quality measure.
|