Preprocessing method and system for multi-road segment data of lane line crowdsourcing data

Abstract

The invention relates to a preprocessing method and system for multi-road segment data of lane line crowdsourcing data. The method comprises the steps that: 1, shape points in a single lane line in the road segment data are sorted according to coordinates in one axial direction; 2, whether the difference between the direction of the shape point pn and the connecting line direction of head and tailshape points exceeds a set threshold value or not is judged; if the difference between the direction of the shape point pn and the connecting line direction of head and tail shape points exceeds theset threshold value, the shape point pn+1 is deleted, otherwise, 1 is added to n, the direction of the shape point pn is the connecting line direction of the shape point pn and the shape point pn+1; and 3, the step 2 is circularly executed until the value of n is the total number of the shape points, and linear fitting optimization is performed on the shape points of each lane line. According to the method, all the shape points in each lane line are sorted according to the collinear relations between the points; abnormal outliers are filtered out; lane line data collected by crowdsourcing collection vehicles are effectively preprocessed; the effect and efficiency of subsequent optimization processing are improved; and therefore, high-precision lane line map data meeting a precision requirement are obtained.

Description

technical field [0001] The present invention relates to the field of high-precision maps, in particular to a method and system for preprocessing multi-road segment data of lane line crowdsourcing data. Background technique [0002] In the field of autonomous driving, in order to accurately control the driving of vehicles, the drawing of high-precision maps is often involved. High-precision maps can be drawn after long-term data collection by expensive surveying and mapping vehicles. However, due to high costs, long collection periods, and slow update The reason is that it is difficult to meet the high freshness requirements of high-precision maps. [0003] Compared with high-precision surveying and mapping vehicles, the cost of crowdsourcing collection vehicles is lower, and it is more suitable for extensive deployment to collect high-quality data and improve the update frequency of high-precision maps. Large and often wrong data points cannot be directly used for optimizat...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the technical problems existing in the prior art, the present invention provides a preprocessing method and system for multi-road segment data of lane line crowdsourcing data, and solves the problem of low accuracy of crowdsourcing data in the prior art

Images