Lane line quality calculation method and device based on lane line curve coefficient and automobile

Abstract

The invention provides a lane line quality calculation method based on a lane line curve coefficient, and the method comprises the steps of storing a new frame of lane line data fused and output by a multi-sensor data fusion device of a vehicle into a container when the new frame of lane line data is received, wherein the lane line data comprises the lane line curve coefficient and a lane line length; when the (n+1)-th frame of lane line data to be stored is received, discarding the first frame of lane line data currently stored in the container so as to enable the maximum cache amount in the container to be n frames of lane line data; when a frame of new lane line data is stored in the container, carrying out standard deviation solution on the curve coefficient of each lane line; taking the smaller upper limit value of the solved standard deviation of the lane line curve coefficients and the preset standard deviation of the lane line curve coefficients, and selecting a final standard deviation of the lane line curve coefficients; and calculating the quality of the lane line based on the final standard deviation of the lane line curve coefficients.

Description

technical field [0001] The invention relates to the technical field of automatic driving perception fusion, in particular to a lane line quality calculation method, device and vehicle based on lane line curve coefficients. Background technique [0002] In the autonomous driving perception fusion technology, lane line fusion is a necessary process to achieve lateral control. The lateral control of self-driving vehicles generally generates a center trajectory line through the lane line, so that the vehicle can drive according to the trajectory to achieve the lateral control effect. When relying on lane lines for lateral control, the effect of lateral control not only depends on the quality of the control algorithm, but also depends on whether the input lane lines are good enough to be used for lateral control. The existing automatic driving system mainly judges whether the lane line can be used for control at the control end, and relies on a single feature directly output fro...

AI Technical Summary

Problems solved by technology

A single feature often represents the quality of a specific characteristic of the lane line, and cannot represent the overall quality of the lane line;

[0005] The characteristics of lane line quality output by different sensors have inherent differences due to different sensor types, and these features are difficult to unify under the same standard when lane line fusion

Images