Lane keeping system and method based on Beidou high-precision positioning

Abstract

The invention provides a lane keeping system based on Beidou high-precision positioning. The lane keeping system comprises a vehicle running state information acquiring module, a BDS receiving end, an upper computer control system, a lower computer control system and a terminal monitoring system, wherein the vehicle running state information acquiring module acquires parameters of vehicle running states from a vehicle; the BDS receiving end receives position information of the vehicle, a map and road information from a Beidou satellite; the upper computer control system acquires the parameters of the vehicle running states, the position information of the vehicle, the map and the road information through a CAN bus, a suitable lane is selected, the vehicle is tracked in real time according to a vehicle tracking algorithm, and the position information of the vehicle is fed back; the lower computer control system acquires the suitable lane and transmits a control command to the vehicle; and the terminal monitoring system communicates with the upper computer control system through a network. The device is directly mounted in the vehicle, a background server or wireless network bandwidth is not required, the cost is low, and cost for data flow is not required to be paid in a later period.

Description

technical field [0001] The invention belongs to the field of Beidou positioning, and in particular relates to a lane keeping system and method based on Beidou high-precision positioning. Background technique [0002] The BeiDou Navigation Satellite System (BDS), which consists of three parts: the space segment, the ground segment and the user segment, can provide all types of users with high-precision, high-reliability positioning, Navigation, timing service, and short-message communication capability. It has preliminary regional navigation, positioning and timing capabilities. The positioning accuracy is 10 meters, the speed measurement accuracy is 0.2 m / s, and the timing accuracy is 10 nanoseconds. [0003] There are several methods in the lane recognition method: the lane path recognition method based on machine learning, but the reliability of the recognition is not high; the method of combining connected area search and road boundary Hough detection, but the threshold v...

AI Technical Summary

Problems solved by technology

[0003] There are several methods in the lane recognition method: the lane path recognition method based on machine learning, but the reliability of the recognition is not high; the method of combining connected area search and road boundary Hough detection, but the threshold value processing method used It is relatively simple, and it is difficult to obtain ideal recognition results in complex road surface environments; based on the dynamic image threshold acquisition method of lane line width, this method can adaptively adjust the recognition threshold to obtain recognition thresholds suitable for various environments, but only in lanes The effect is better when the lines are clear and undamaged, but there are also great limitations

[0004] There are several methods in the lane tracking method: the navigation parameter extraction method, which effectively reduces the amount of image data processing, but the extracted control parameters are difficult to adapt to changes in road curvature; there is also a method using preview tracking controller, It can obtain control parameters that adapt to changes in road curvature in real time, but the calculation data processing volume of road curvature is large, which is not conducive to improving real-time performance; there is also a method based on the output of the kinematic controller as the input method of the dynamic controller, but lacks control information output and feedback

Images