Vehicle positioning method and system

Abstract

Embodiments of the present invention provide a vehicle positioning method and system. The method includes: synchronously acquiring output information of an angle sensor installed in a vehicle and an output image of a camera; calculating the difference between the output image of the current frame and the output image of the previous frame If the lateral movement amount is greater than the preset threshold, the rotation angle obtained based on the output information of the angle sensor is adopted; otherwise, the rotation angle obtained based on the lateral movement amount and / or the longitudinal movement amount is adopted. Rotation angle; the movement of the camera is calculated based on the lateral movement amount, the longitudinal movement amount and the rotation angle to position the vehicle. The invention combines the motion characteristics of the vehicle, amplifies the rotation angle of the vehicle through the camera, and obtains a more accurate vehicle rotation angle in combination with the angle sensor, so as to solve the problem of positioning failure caused by the accumulated angle error, and then solve the positioning cost in the scene of GNSS signal failure. high and low positioning accuracy.

Description

technical field [0001] The present invention relates to the technical field of vehicles, and more particularly, to a vehicle positioning method and system. Background technique [0002] High-precision positioning is a necessary condition for realizing unmanned driving. At present, the Global Positioning Satellite System (GNSS) is the most used vehicle positioning system, but for indoor parking lots, tunnels and other scenarios, the GNSS signal is interfered, which seriously affects the positioning accuracy. In this case, it is often necessary to rely on high-cost lidar or high-cost fiber optic gyroscopes and quartz accelerometers for positioning. Compared with the above solutions, the solution based on visual positioning has lower cost and is easier to be widely used. The movement of the vehicle can be divided into two parts: rotation and translation. When the rotation angle is not calculated accurately, it will often cause a large positioning error, such as figure 1 sho...

AI Technical Summary

Problems solved by technology

When the calculation of the rotation angle is inaccurate, it will often cause a large positioning error, such as figure 1 shown

When calculating the rotation angle, due to the small rotation angle of the single frame of the vehicle, the vision-based positioning has a large relative error when calculating the rotation angle of the front and rear frames

Images