Lane tracking method and system for autonomous vehicles

Abstract

A lane tracking method and system for self-driving vehicles, the processing unit estimates the yaw rate and lateral acceleration according to the measurement results of the inertial measurement unit and meets the confidence level, and the pre-stored and for the vehicle driving in the lane at the current time The current position data estimated at the point, estimate and store the future position data of the vehicle at the future unit time point, and then estimate the vehicle from the historical time The total longitudinal displacement, the total lateral displacement and the total azimuth offset from the point to the future unit time point, as well as the reference lane line data pre-stored and generated based on the image of the sensed lane at the historical time point and using coordinate transformation The future lane line data corresponding to the future unit time point is estimated, so that the update frequency of the lane line data for lateral track tracking can be increased.

Description

technical field [0001] The present invention relates to the control of automatic driving vehicles, in particular to a lane tracking method and system for automatic driving vehicles. Background technique [0002] According to the L1-L5 automatic driving technology level formulated by the International Society of Automatic Machine Engineers (SAE), for the higher level of automatic driving technology, the robustness and reliability of the active control system of the automatic driving vehicle will face even greater challenges. high challenge. [0003] However, in view of the current camera’s limitation on the processing time required for the captured image from processing to output, the update frequency of lane detection information (for example, 10-20Hz) is too low, and the output information of lane detection is common. The time asynchronous problem with the vehicle motion control, for example, the lane line information provided to the ECU of the vehicle control system is us...

AI Technical Summary

Problems solved by technology

On the other hand, if the control frequency is too low due to the limited update frequency of lane sensing information, for example, this may lead to a decrease in the resolution and accuracy of control commands, especially for high curvature road sections or lateral deviation. When the moving speed is large, the vehicle trajectory tracking effect is not good

[0004] In addition, existing roads often have lane markings that are stained, blurred, or have abnormal color contrast of lane markings due to light changes, or even do not exist at all (such as at intersections), which may lead to lane detection. The image recognition error of the detection module or the lane detection function is abnormal, or even the lane line detection function cannot be provided at all.

Images