Preprocessing method, device, equipment and medium for lane line perception data

Abstract

The present application discloses a method, device, device and medium for preprocessing lane line perception data, and relates to a lane line detection technology. The specific implementation scheme is: obtain the high-precision data of the lane line of the lane where the automatic driving device is located at the current moment from the high-precision map; use the high-precision data of the lane line as the prior data, and use the perception data of the lane line sensed at the current moment as the observation data, The Kalman filter is used to estimate the actual lane line data to obtain the lane line target data at the current moment. The embodiment of the present application uses the Kalman filter combined with the high-precision map to perform state estimation on the actual lane line data, which can filter out erroneous data in the perception data, generate accurate and smooth lane line target data, and avoid vehicle position and attitude solutions. It can provide more stable and smooth position and attitude data for downstream modules to improve the safety of automatic driving. This solution can be used for autonomous driving, especially in the field of autonomous parking.

Description

technical field [0001] The present application relates to the technical field of automatic driving, in particular to a lane line detection technology, and in particular to a preprocessing method, device, device and medium for lane line perception data. Background technique [0002] In recent years, automatic driving technology has developed rapidly, and the detection of lane lines is an important part of the automatic driving system. The detection of lane lines is to separate the lane lines from other data in the image according to the color, shape and texture of the lane lines in the video image obtained by the vehicle camera, so as to obtain the direction of the lane lines and the direction of the vehicle relative to the lane. information about the position of the line. [0003] Due to the complex and changeable actual driving environment, whether the lane lines obtained by traditional statistical learning methods or deep learning methods often have lateral intercept jump...

AI Technical Summary

Problems solved by technology

[0003] Due to the complex and changeable actual driving environment, no matter whether it is based on the traditional statistical learning method or the deep learning method, the lane lines often have lateral intercept jumps, missed detections or false detections, and even in some scenarios. It may also cause the vehicle to be positioned incorrectly, so it is difficult to directly use it in subsequent positioning algorithms to obtain the vehicle's position and attitude, resulting in the inability of the vehicle to perform safe automatic driving

Images