Method and device for generating parking model, electronic device and storage medium

Abstract

The application discloses a parking model generation method and device, an electronic device and a storage medium, and relates to the field of autonomous parking. The specific implementation plan is: obtain multiple parking sample sets; for each parking sample set, construct a parking cruising space for the target vehicle according to the parking sample set, and extract the parking cruising space corresponding to the parking sample set in the parking cruising space. The first parking track; identify the abnormal position on each first parking track, and delete the parking driving data of the abnormal position from the corresponding parking sample set, so as to obtain the parking model used in the parking sample set Target sample data; use the target sample data to train the parking model to generate the target parking model, solve the problem of inaccurate data when building the parking model, and effectively denoise the parking data for building the target parking model, In this way, a parking model with more accurate results and higher reliability can be constructed.

Description

technical field [0001] The present application relates to the technical field of data processing, in particular to the technical field of autonomous parking, and in particular to a method and device for generating a parking model, electronic equipment and a storage medium. Background technique [0002] With the development of society and the improvement of people's living standards, people's demand for automatic parking is increasing, especially in scenes with fixed parking spaces such as homes and companies, and they hope that automatic parking can bring them the freedom of the last mile . [0003] In related technologies, when training the parking model, the data collected in real time is usually used, that is, there is no need to collect high-precision maps in advance, but the road condition information collected in real time is directly used as samples for model training. However, since the data collected in real time is still It is easy to contain special road conditio...

AI Technical Summary

Problems solved by technology

[0003] In related technologies, when training the parking model, the data collected in real time is usually used, that is, there is no need to collect high-precision maps in advance, but the road condition information collected in real time is directly used as samples for model training. However, since the data collected in real time is still It is easy to contain special road condition information and driving trajectories, such as avoiding temporary pedestrians and other vehicles, etc. Therefore, there are prone to data anomalies, resulting in the learned control strategy cannot guarantee the safety of autonomous parking, affecting user experience

Images